Re: [freenet-dev] Installer file name
Juiceman skrev: > Currently we distribute our installer using a name such as > FreenetInstaller-1223.exe but we call it 0.7.5 I suggest we use the > nomenclature FreenetInstaller-0.7.5.1223.exe that is more in line with > what appears to be standard amongst many applications. > > Thoughts? I think having both major and minor version number in the file name is a good idea, just as you suggest. I'd like to stress once again that https://bugs.freenetproject.org/view.php?id=3193 (Building the wininstaller: src_freenetinstaller/FreenetInstaller_Include_Info.inc is not updated with the latest build info) should be fixed ASAP. Until it is fixed, the wininstaller will show a wrong version number in the main GUI. - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Updating helper executables on Windows
Juiceman skrev: > On Tue, Jul 14, 2009 at 5:45 PM, Juiceman wrote: >> On Tue, Jul 14, 2009 at 8:46 AM, Zero3 wrote: >>> Juiceman skrev: >>>> On Mon, Jul 13, 2009 at 8:37 PM, Zero3 wrote: >>>>> Juiceman skrev: >>>>>> On Mon, Jul 13, 2009 at 10:59 AM, Zero3 wrote: >>>>>>> Juiceman skrev: >>>>>>>> I'm working on the update.cmd script handling these binaries... it >>>>>>>> seems the .sha1 of all of these files is blank on the website. The >>>>>>>> wrappers, start.exe, stop.exe and the freenetlauncher.exe. I'll need >>>>>>>> that fixed please to continue my work ;-) >>>>>>> Cool! Remember to update freenettray.exe as well. > > > I think I have finished update.cmd to handle all of the binaries. > I have these sections of the script bypassed until non-blank .sha1's > of the files on the website can be generated and Zero3 can finish the > freenettray.exe utility. Once those are done, I can fully test those > sections of the script and wire them in for final release. > We don't necessarily need to wait for the freenettray. The sha1's > will let me do most of the work that remains... > > Balls in your court gentlemen. =) Sweeet! I'm currently doing *major* update of the translation system in the wininstaller (mostly to simlpify the translation process for translators and reduce the number of strings and their lengths by reusing common ones). The service account reworking and tray manager are already done (but needs testing, if anyone bothers!). - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Beta-testers for Facebook-plugin
Luke771 skrev: > Jonas Bengtsson wrote: >> As part of my master-thesis "Growing secure peer-to-peer networks" I >> have created a facebook plugin that allows a user to share his darknet >> node reference with facebook friends > > bad idea IMHO I think it is a great idea really (and one often requested, nonetheless). The whole idea of the darknet is that you exchange references with your friends, and so far Freenet has absolutely failed at making people do just that. A plugin that somehow imports/exchanges/syncs/... friends between Facebook (probably the greatest WoT on earth) and Freenet might just help build the darknet we need. - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Experiences writing a plugin
Evan Daniel skrev: > Having not written much actual Freenet code before, I'm learning a lot > about how Freenet works in the process -- which is harder than it has > any reason to be. Why? NOTHING IS DOCUMENTED. [snip] > If this were an isolated incident, it wouldn't matter much. It isn't. > It is the norm for Freenet. For a platform whose primary impediment > to wider adoption (IMO, of course) is a lack of things to do with it, > rather than a lack of underlying functionality, this is a problem. I > haven't tracked it, but I wouldn't be surprised if I've spent nearly > as much time trying to figure out how the plugin API works (or even > which classes it consists of) as I have actually writing code. [snip] > At this point, I think I have a much better understanding of why > Freenet has so little software that makes use of it, despite the fact > that Freenet itself seems to work fairly well. I completely agree. I've been pulling my hair over similar issues before as well. The closest thing to documentation was for me the Wiki and/or simply askin toad about what I needed to know. I guess the fact that the Freenet core is ever-changing has a lot to do with it. - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] System tray status was Re: Updating helper executables on Windows
Matthew Toseland skrev: > On Monday 13 July 2009 15:59:21 Zero3 wrote: >> Juiceman skrev: >>> I'm working on the update.cmd script handling these binaries... it >>> seems the .sha1 of all of these files is blank on the website. The >>> wrappers, start.exe, stop.exe and the freenetlauncher.exe. I'll need >>> that fixed please to continue my work ;-) >> Cool! Remember to update freenettray.exe as well. > > Ummm, so freenettray.exe is now shipping with new installs? Or...? > > What is the status of the freenet tray icon on Windows? It's in the beta branch (you are most likely building from master). Status is done. It could use more testing and usability feedback though. I'm posting the latest beta of the installer (which includes it, among other things) to this list in a moment. - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] No-Freenet-user installer was Re: install problem XP SP2 java6update13
Matthew Toseland skrev: > On Tuesday 07 July 2009 22:50:04 Zero3 wrote: >> Matthew Toseland skrev: >>>> It would be a great help if you could test the latest beta version of >>>> the Windows installer which includes this change. The source is >>>> available in our git repository, but I figure out probably prefer a >>>> compiled version, so I've put one online at: >>>> >>>> http://privat.zero3.dk/FreenetInstaller_Beta.exe >>> When do you expect to be able to commit this to the app-wininstaller repo? >> It is already there, in the beta branch. But due to internal >> re-organizing and the depency of an updated version of the update.cmd >> script (Juiceman is working on this), it shouldn't be merged back into >> master just yet. More testing would be nice as well. >> >> One of the 3 persons I've sent the beta link to has finally responded, >> so I'm currently investigating this issue. The wrapper's CreateService() >> still fails for him though, even without our custom user - so I've >> sent him details of how we create the service and await his debugging >> results. >> >> The main problem in tracking down this bug is that it only occurs to a >> very few users (those 3 so far) and there doesn't appear to be anything >> in common between them besides the fact that they run Vista. I cannot >> reproduce it on my own Vista setup either... > > Progress? Nop - he still hasn't replied to my last mail. I suspect that this might be a wrapper bug instead since the latest version (which I've included in the latest beta) fixes at least one crash. - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
[freenet-dev] Help wanted: An improved 16x16 px Windows tray icon
The new Windows tray manager is right now using a simple down-scaled version of the Freenet bunny. It works okay, but it isn't optimal. If anyone can fix up a better icon, please feel free to do so. The current ones are available here: Icon for "active": http://github.com/freenet/wininstaller-staging/raw/7976746de8f198e04f39c9a7b0f0e740c559a9d4/src_freenetinstaller/files_install/freenet.ico Icon for "inactive": http://github.com/freenet/wininstaller-staging/raw/7976746de8f198e04f39c9a7b0f0e740c559a9d4/src_freenetinstaller/files_install/freenetoffline.ico Reference icon (bigger sizes available inside): http://github.com/freenet/wininstaller-staging/raw/8861cd5be416ead97ffe68ff9fdaa76c285d7fb9/src_freenetinstaller/FreenetInstaller_Icon.ico - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
[freenet-dev] Testing wanted: New Windows installer beta
Hi all I've released a new beta of the Windows installer. Please feel free to test away! Feedback of all kinds wanted. Source: http://github.com/freenet/wininstaller-staging/tree/beta Binary: http://privat.zero3.dk/FreenetInstaller_Beta.exe This beta covers the following reports on the bugtracker: https://bugs.freenetproject.org/view.php?id=3219 (System tray icon applet for Windows) https://bugs.freenetproject.org/view.php?id=1709 (The installer bundles NETUSER.exe which we don't have any control over.) https://bugs.freenetproject.org/view.php?id=3080 (Wininstaller does not remove the user profile on Vista) https://bugs.freenetproject.org/view.php?id=3195 (Fails to install, always system error 1069) https://bugs.freenetproject.org/view.php?id=3192 (Find replacements for netuser.exe and Ntrights.exe) https://bugs.freenetproject.org/view.php?id=3216 (Make a version without the custom user) https://bugs.freenetproject.org/view.php?id=3303 (Pass on incognito flag to fproxy when using an incognito-enabled browser) https://bugs.freenetproject.org/view.php?id=3104 (Usability: let the user choose a browser and/or install one in the installer) Furthermore, the translation system has been reworked, making it easier and quicker for translators to localize the Installer and keeping their translations updated. I will send out a separate mail about this soon. - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Installer file name
Matthew Toseland skrev: > On Wednesday 22 July 2009 15:41:54 bren...@artvote.com wrote: >> Just curious, whom exactly does the 4 digit number benefit? Do users care >> about this number? And if so why? (Sorry if these are dumb questions. Just >> trying to wrap my head around the issue:) >> -Brendan > > Basically the problem is the CDN we use (Google Code). A filename should > uniquely identify its contents, when we change the contents we should change > the filename. > > Is FreenetInstaller-0.7.5.1224.exe a problem? IMHO an installer should always contain the version number in its name so that there is no doubt about which version it contains. The name sounds fine to me. - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Need definition of fproxy language codes
Daniel Cheng skrev:
> On Mon, Jul 6, 2009 at 8:58 AM, Zero3 wrote:
>> Reposting:
>>
>> bo-le skrev:
>>> Am Dienstag, 16. Juni 2009 21:40:53 schrieb Zero3:
>>>> Matthew Toseland skrev:
>>>>> On Sunday 14 June 2009 13:11:40 Zero3 wrote:
>>>>>> Matthew Toseland skrev:
>>>>>> This value is also passed on to the node via "node.l10n=Deutsch"
>>>>>> (example for German) in freenet.ini. (I don't think that specifying a
>>>>>> language by the localized name is ideal, but that seems to be how the
>>>>>> node wants it. I *did* ask if this could be changed to standardized
>>>>>> language IDs a while ago...)
>>>>> IIRC both work.
>>>> Which kind of IDs does it accept? I'd really like to switch over to that
>>>> instead.
>>> freenet.l10n.L10n.java shows you the strings:
>>> /** @see "http://www.omniglot.com/language/names.htm"; */
>>> public enum LANGUAGE {
>>> ENGLISH("en", "English", "eng"),
>>> SPANISH("es", "Español", "spa"),
>>> DANISH("da", "Dansk", "dan"),
>>> GERMAN("de", "Deutsch", "deu"),
>>> FINNISH("fi", "Suomi", "fin"),
>>> FRENCH("fr", "Français", "fra"),
>>> ITALIAN("it", "Italiano", "ita"),
>>> NORWEGIAN("no", "Norsk", "nor"),
>>> POLISH("pl", "Polski", "pol"),
>>> SWEDISH("se", "Svenska", "svk"),
>>> CHINESE("zh-cn", "中文(简体)", "chn"),
>>> CHINESE_TAIWAN("zh-tw", "中文(繁體)", "zh-tw"),
>>> UNLISTED("unlisted", "unlisted", "unlisted");
>>>
>>> any string listed here can be used.
>> Cool. But which standards do these follow? ,
>> and (though zh-tw seems wrong then)?
>>
>
> Grad you ask about "zh-tw" :)
>
> The last field is supposed to be countries, or "Regional Authority".
> Here we have 3 problems:
>
> 1) Language does not match "Regional Authority"
> For example, Spanish and English are used in USA
> English is used in many commonwealth region.
> Some minority language never have an official position
> in ANY region.
>
> 2) Spoken Language and Written Language never match
> 中文(繁體) and 中文(简体) are different writing script for the
> same (spoken) language family.
>
> 3) Some "Authority" are not globally recognized .
> The "tw" in "zh-tw" stand for "Taiwan". Do I have to explain more?
>
> In short, we simply don't have any choice standard there.
Ugh. So what the heck do I do?
Right now, the wininstaller simply passes on the localized name of the
language to fproxy, but that probably won't match most of the time. How
should I negotiate with fproxy about which language to use? (... because
another variation of French (for example), should still be a better
choice than falling back to English).
IMHO this is a complete mess, and we are probably better off letting
fproxy itself handle the language selection based on OS locale or user
choice or whatever. Doesn't fproxy pull the OS locale and pre-select
according to that anyway?
- Zero3
___
Devl mailing list
Devl@freenetproject.org
http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
[freenet-dev] Default plugins, freenet.ini defaults and wrapper.conf defaults
Are the following defaults (the ones I use in the Windows installer) still up-to-date? Anything that needs added/removed/updated? [Plugins] JSTUN KeyExplorer ThawIndexBrowser UPnP XMLLibrarian [freenet.ini] fproxy.port= fcp.port= pluginmanager.loadplugin=JSTUN;KeyExplorer;ThawIndexBrowser;UPnP;XMLLibrarian node.updater.autoupdate=true node.l10n= [wrapper.conf] wrapper.java.command=java wrapper.working.dir=../ wrapper.java.mainclass=freenet.node.NodeStarter wrapper.java.classpath.1=freenet.jar wrapper.java.classpath.2=freenet-ext.jar wrapper.java.library.path.1=lib wrapper.java.initmemory=60 wrapper.java.maxmemory=192 wrapper.java.additional.1=-Dnetworkaddress.cache.ttl=0 wrapper.java.additional.2=-Dnetworkaddress.cache.negative.ttl=0 wrapper.java.additional.3=-enableassertions:freenet wrapper.app.parameter.1=freenet.ini wrapper.console.format=PM wrapper.console.loglevel=INFO wrapper.logfile=wrapper.log wrapper.logfile.format=LPTM wrapper.logfile.loglevel=INFO wrapper.logfile.maxsize=2M wrapper.logfile.maxfiles=3 wrapper.syslog.loglevel=NONE wrapper.console.title=Freenet wrapper.jvm_exit.timeout=120 wrapper.restart.reload_configuration=TRUE wrapper.filter.trigger.1=java.lang.OutOfMemoryError wrapper.filter.action.1=RESTART wrapper.ntservice.description=Freenet background service wrapper.ntservice.dependency.1= wrapper.ntservice.starttype=AUTO_START wrapper.ntservice.interactive=false wrapper.ntservice.process_priority=BELOW_NORMAL wrapper.ntservice.name= wrapper.ntservice.displayname= wrapper.ntservice.account= - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Installer file name
Matthew Toseland skrev: > On Saturday 25 July 2009 20:48:05 Zero3 wrote: >> Matthew Toseland skrev: >>> On Wednesday 22 July 2009 15:41:54 bren...@artvote.com wrote: >>>> Just curious, whom exactly does the 4 digit number benefit? Do users care >>>> about this number? And if so why? (Sorry if these are dumb questions. Just >>>> trying to wrap my head around the issue:) >>>> -Brendan >>> Basically the problem is the CDN we use (Google Code). A filename should >>> uniquely identify its contents, when we change the contents we should >>> change the filename. >>> >>> Is FreenetInstaller-0.7.5.1224.exe a problem? >> IMHO an installer should always contain the version number in its name >> so that there is no doubt about which version it contains. > > Yes, except possibly for download-the-installer-from-freenet? If it's fetched > from the web interface we could do it there too though... I personally dislike when you have SomethingInstaller.exe and no idea which version it contains (and if it is up to date according to whatever latest version the website announces). Or even worse, 2x SomethingInstaller.exe with different sizes and no idea which is latest. It's not a dealbreaker to me though. Whatever works out for you. Distributing it over Freenet would be cool. - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] status
Evan Daniel skrev: > On Tue, Jul 28, 2009 at 3:26 PM, Florent > Daignière wrote: >> * Matthew Toseland [2009-07-28 20:03:42]: >>> So we need to deal with the bug tracker. But it will take several days work >>> (= approximately the cost to both FPI and Ian of emu for 1.5 months). Is >>> this really a high priority? IMHO losing our existing bugs database will >>> cost significant work in the medium to long term, hence the need to migrate >>> data... >> Last time people talked about it, only you and xor where objecting to >> trashing the bug database altogether and switching to another bug >> tracker (possibly hosted by someone else). > > As someone who has submitted a number of both feature requests and > bugs to the database, I would be right annoyed if they were simply > dropped without any plan to keep track of and address them. That > said, I have no particular attachment to the current software, and no > objection to changing to something else if would improve things. Ditto. I don't think we are using our current bugtracker very effectively right now, but simply trashing everything seems like a bad idea to me too. However, if we are migrating anyway, it might be a good time to stop up and actually clean up the bugtracker (this was suggested earlier too). There are *tons* of outdated bugs in there and very few people that actually know enough about Freenet to do the cleaning. - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Default plugins, freenet.ini defaults and wrapper.conf defaults
Matthew Toseland skrev: > On Saturday 25 July 2009 21:09:14 Zero3 wrote: >> [freenet.ini] >> fproxy.port= >> fcp.port= >> pluginmanager.loadplugin=JSTUN;KeyExplorer;ThawIndexBrowser;UPnP;XMLLibrarian >> >> >> node.updater.autoupdate=true > > No, node.updater.enabled=true, but autoupdate OFF by default, the wizard will > ask. Isn't that default anyway? Is there any reason to explicitly specify it? I'll remove the autoupdate line then. - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Freenet blog plugin?
Matthew Toseland skrev: > It has been pointed out that a minimal blog engine can be written in approx > 22KB of php - around 800 lines of code at most. I suspect that given a > template I could probably put together a blog plugin in a few days. This > would integrate with Freetalk for comments and for announcing the site > initially. It should make it easier to contribute content to Freenet, > eliminating the need to get Thingamablog working, etc. Thoughts? IMHO it would be a better idea to have someone less experienced with Freenet development work on and maintain such (with your mentoring). A project like this seems like a great opportunity to get a new developer into working with Freenet. Given that such person is available, of course. - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Hello Freenet (& design feedback)
Matthew Toseland skrev: > I have deployed the current website redesign, which is essentially Dieppe's > work with much feedback from here. It is working, the only problem is the > translation page is empty: > http://freenetproject.org/translation.html > > Should we remove it for now? > > IMHO the new site is an improvement over the old one, especially with the big > download button. However I'm sure we could further improve it. *claps* I think it is a great improvement! (with room for optimization, as always) The screenshot is still worse than awful though. - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Freenet 0.7 build 1226
Matthew Toseland skrev: > - Set the datastore initially to a 10MB RAM-only store, set it to salted-hash > in the post-install wizard when the size is chosen, automatically migrate the > data. Avoids a lot of unnecessary disk accesses on new installs. I've adjusted the disk requirements in the beta branch of the Windows installer to reflect this. - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Freenet 0.7 build 1226
Matthew Toseland skrev: > On Friday 31 July 2009 23:01:16 Zero3 wrote: >> Matthew Toseland skrev: >>> - Set the datastore initially to a 10MB RAM-only store, set it to >>> salted-hash in the post-install wizard when the size is chosen, >>> automatically migrate the data. Avoids a lot of unnecessary disk accesses >>> on new installs. >> I've adjusted the disk requirements in the beta branch of the Windows >> installer to reflect this. > > I'm not sure they have changed? The last requirement I knew of was a 100 MB datastore created at first node service start. Since the wizard now handles that, I've changed the wininstaller to only require 10 MB (guesstimated install size) + buffer to not fill up the install drive (384 MB atm.). - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Freenet 0.7 build 1226
Matthew Toseland skrev: > On Sunday 02 August 2009 13:40:12 Zero3 wrote: >> Matthew Toseland skrev: >>> On Friday 31 July 2009 23:01:16 Zero3 wrote: >>>> Matthew Toseland skrev: >>>>> - Set the datastore initially to a 10MB RAM-only store, set it to >>>>> salted-hash in the post-install wizard when the size is chosen, >>>>> automatically migrate the data. Avoids a lot of unnecessary disk accesses >>>>> on new installs. >>>> I've adjusted the disk requirements in the beta branch of the Windows >>>> installer to reflect this. >>> I'm not sure they have changed? >> The last requirement I knew of was a 100 MB datastore created at first >> node service start. Since the wizard now handles that, I've changed the >> wininstaller to only require 10 MB (guesstimated install size) + buffer >> to not fill up the install drive (384 MB atm.). > > The wizard won't set it smaller than 256MB. Aaargh. I'll repatch... - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Updating helper executables on Windows
Juiceman skrev: >> Sorry, these are uploaded manually at the moment. The sha1's have been fixed >> for now. > > Thanks! How are you doing on this? And do you need any more info about shutting down tray managers or other stuff? Anything? - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Default plugins, freenet.ini defaults and wrapper.conf defaults
Zero3 skrev: > Matthew Toseland skrev: >> On Saturday 25 July 2009 21:09:14 Zero3 wrote: >>> [freenet.ini] >>> fproxy.port= >>> fcp.port= >>> pluginmanager.loadplugin=JSTUN;KeyExplorer;ThawIndexBrowser;UPnP;XMLLibrarian >>> >>> >>> node.updater.autoupdate=true >> No, node.updater.enabled=true, but autoupdate OFF by default, the wizard >> will ask. > > Isn't that default anyway? Is there any reason to explicitly specify it? > I'll remove the autoupdate line then. *bump* - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Need definition of fproxy language codes
Zero3 skrev:
> Daniel Cheng skrev:
>> On Mon, Jul 6, 2009 at 8:58 AM, Zero3 wrote:
>>> Reposting:
>>>
>>> bo-le skrev:
>>>> Am Dienstag, 16. Juni 2009 21:40:53 schrieb Zero3:
>>>>> Matthew Toseland skrev:
>>>>>> On Sunday 14 June 2009 13:11:40 Zero3 wrote:
>>>>>>> Matthew Toseland skrev:
>>>>>>> This value is also passed on to the node via "node.l10n=Deutsch"
>>>>>>> (example for German) in freenet.ini. (I don't think that specifying a
>>>>>>> language by the localized name is ideal, but that seems to be how the
>>>>>>> node wants it. I *did* ask if this could be changed to standardized
>>>>>>> language IDs a while ago...)
>>>>>> IIRC both work.
>>>>> Which kind of IDs does it accept? I'd really like to switch over to that
>>>>> instead.
>>>> freenet.l10n.L10n.java shows you the strings:
>>>> /** @see "http://www.omniglot.com/language/names.htm"; */
>>>> public enum LANGUAGE {
>>>> ENGLISH("en", "English", "eng"),
>>>> SPANISH("es", "Español", "spa"),
>>>> DANISH("da", "Dansk", "dan"),
>>>> GERMAN("de", "Deutsch", "deu"),
>>>> FINNISH("fi", "Suomi", "fin"),
>>>> FRENCH("fr", "Français", "fra"),
>>>> ITALIAN("it", "Italiano", "ita"),
>>>> NORWEGIAN("no", "Norsk", "nor"),
>>>> POLISH("pl", "Polski", "pol"),
>>>> SWEDISH("se", "Svenska", "svk"),
>>>> CHINESE("zh-cn", "中文(简体)", "chn"),
>>>> CHINESE_TAIWAN("zh-tw", "中文(繁體)", "zh-tw"),
>>>> UNLISTED("unlisted", "unlisted", "unlisted");
>>>>
>>>> any string listed here can be used.
>>> Cool. But which standards do these follow? ,
>>> and (though zh-tw seems wrong then)?
>>>
>> Grad you ask about "zh-tw" :)
>>
>> The last field is supposed to be countries, or "Regional Authority".
>> Here we have 3 problems:
>>
>> 1) Language does not match "Regional Authority"
>> For example, Spanish and English are used in USA
>> English is used in many commonwealth region.
>> Some minority language never have an official position
>> in ANY region.
>>
>> 2) Spoken Language and Written Language never match
>> 中文(繁體) and 中文(简体) are different writing script for the
>> same (spoken) language family.
>>
>> 3) Some "Authority" are not globally recognized .
>> The "tw" in "zh-tw" stand for "Taiwan". Do I have to explain more?
>>
>> In short, we simply don't have any choice standard there.
>
> Ugh. So what the heck do I do?
>
> Right now, the wininstaller simply passes on the localized name of the
> language to fproxy, but that probably won't match most of the time. How
> should I negotiate with fproxy about which language to use? (... because
> another variation of French (for example), should still be a better
> choice than falling back to English).
>
> IMHO this is a complete mess, and we are probably better off letting
> fproxy itself handle the language selection based on OS locale or user
> choice or whatever. Doesn't fproxy pull the OS locale and pre-select
> according to that anyway?
*bump*
- Zero3
___
Devl mailing list
Devl@freenetproject.org
http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Some statistics
Matthew Toseland skrev: > Also the uninstall survey has dried up, there are very few responses now, too > few to be useful. Maybe we should always show it if opennet was enabled? Or > would even that be too much? Please don't. It is very hostile (usability wise) to automatically pop up these kind of surveys. Not even speaking of the fact that we effectively would stab uninstalling users in the back by automatically contacting freenetproject.org (which *MOST LIKELY* will be monitored in hostile regimes). - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] How to start a new translation?
Alex Pyattaev skrev: > Hi there! > I'd like to make a russian translation for freenet program (it seems it > will be very popular here some time soon). > I just want to know how to organize my work so that it is not lost in vain. > Thanks. Hi Alex If you want to translate the Windows Installer as well, please feel free to do so. Take a look at the translation guide in git, or via github at: http://github.com/freenet/wininstaller-staging/raw/beta/src_translationhelper/TranslationTemplate.inc If in doubt, you can look at one of the existing translations to see how it is done: http://github.com/freenet/wininstaller-staging/raw/beta/src_translationhelper/Include_Lang_da.inc - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Some statistics
Evan Daniel skrev:
> On Fri, Aug 7, 2009 at 9:13 AM, Zero3 wrote:
>> Matthew Toseland skrev:
>>> Also the uninstall survey has dried up, there are very few responses now,
>>> too few to be useful. Maybe we should always show it if opennet was
>>> enabled? Or would even that be too much?
>> Please don't. It is very hostile (usability wise) to automatically pop
>> up these kind of surveys. Not even speaking of the fact that we
>> effectively would stab uninstalling users in the back by automatically
>> contacting freenetproject.org (which *MOST LIKELY* will be monitored in
>> hostile regimes).
>
> If the user was running opennet in a hostile regime, they already shot
> themselves in the foot. Loading that web page won't make things any
> worse.
Surely there is no point in nailing the coffin an extra time by
connecting to something as obviously Freenet-ish as our official home page?
I don't think we should assume that because opennet was enabled, the
user is caught anyway, and thereby thinking that we might as well use
the opportunity to ask for his uninstallation feedback ("Thanks. You
just gave me 10 years in prison. Best regards, Mr. Wong" ;)).
- Zero3
___
Devl mailing list
Devl@freenetproject.org
http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] 0.8 features? was Re: F2F web proxy???
Evan Daniel skrev: > Here are some specific bugs that I think might make sense to target > for 0.8. Most are UI-related, and I think all provide a good payoff > to work required ratio. > > 3335: HTMLEncoder filtering. There may be a security issue here; then > again, there might not. I would be a lot happier if someone qualified > could say whether there was, and if the code in question referenced > the set of specs it was trying to meet. > 3362: Uploading a folder through fproxy doesn't work. This is bad. > 3365: Freenet should give useful errors about malformed URLs, and > correct them when it can do so easily. > 3295: Percent encoding is horribly broken. Basically any filename > that includes % or other encoded symbols is likely to break something > at some point or other. > 3343: Make the default download directory more intelligent. > 3342: Specify download dir per file queued. Simple, and it's been > requested on Frost. > 3214: Uninstall survey should be functional before we do the next release. > 2692: Can we please either actually fix the peers disconnected bug, or > both make it dismissable and change the wording so we don't keep > getting dozens of duplicate reports? > 3314: Detailed logging thresholds silently break if there are spaces; > makes debugging annoying. > 3313: More intelligent disk usage defaults would be good. > 3340: Reports of headless nodes stalling on startup for lack of > entropy are not uncommon. This would provide an effective if > inelegant fix. > 3273: Remove all completed uploads / downloads is a feature I've seen > requested several times. > 3143: Useful error messages are always a good thing. (Plugin loading > from URLs in this case.) > 3274: Download queue page displays weirdly on first visit. I have a top-5 wishlist as well. I think these all would provide great advantages/improvements compared to the time their implementations would take (or simply fix stupid bugs that should have been fixed a lng time ago): https://bugs.freenetproject.org/view.php?id=1343 (Darknet: short references / out of band password verification) [friend codes] https://bugs.freenetproject.org/view.php?id=2573 (Config of memory limits broken??) https://bugs.freenetproject.org/view.php?id=2026 (Better icons for *nix) https://bugs.freenetproject.org/view.php?id=1720 (Automatic calibration of memory usage?) [automatically adjust memory usage depending on how much free the system has] https://bugs.freenetproject.org/view.php?id=2370 (Lots of users confused by Freenet telling them that it might not be port forwarded) -Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Some statistics
Evan Daniel skrev:
> On Fri, Aug 7, 2009 at 9:29 AM, Zero3 wrote:
>> Evan Daniel skrev:
>>> On Fri, Aug 7, 2009 at 9:13 AM, Zero3 wrote:
>>>> Matthew Toseland skrev:
>>>>> Also the uninstall survey has dried up, there are very few responses
>>>>> now, too few to be useful. Maybe we should always show it if opennet was
>>>>> enabled? Or would even that be too much?
>>>> Please don't. It is very hostile (usability wise) to automatically pop
>>>> up these kind of surveys. Not even speaking of the fact that we
>>>> effectively would stab uninstalling users in the back by automatically
>>>> contacting freenetproject.org (which *MOST LIKELY* will be monitored in
>>>> hostile regimes).
>>> If the user was running opennet in a hostile regime, they already shot
>>> themselves in the foot. Loading that web page won't make things any
>>> worse.
>> Surely there is no point in nailing the coffin an extra time by connecting
>> to something as obviously Freenet-ish as our official home page?
>>
>> I don't think we should assume that because opennet was enabled, the user is
>> caught anyway, and thereby thinking that we might as well use the
>> opportunity to ask for his uninstallation feedback ("Thanks. You just gave
>> me 10 years in prison. Best regards, Mr. Wong" ;)).
>
> If he turned on opennet, then it downloaded the seednodes list from
> that server, right? I suppose there's some chance he installed it in
> a safe regime, moved to a hostile regime that monitors the web server
> but not the seednodes, didn't turn off opennet, and is now
> uninstalling -- but that seems like a rather unlikely case.
We used to do that in the online installers. AFAIK we only use bundle
installers now? (We do for Windows, at least. Which are most of the
installations). In the bundle installers, the seednodes file is included
If the user got the installer straight off our homepage though,
obviously it won't matter, no :). The whole point of switching to bundle
installers are to allow people in hostile regimens to actually install
Freenet by getting the installer from a safe source.
> Besides, if you're uninstalling, you're not doing anything illegal any
> more. I don't think I've heard about people disappearing just for
> visiting a single web page, even in rather oppressive regimes. The
> problem is running Freenet, not visiting the web site (though that
> might get you noticed).
Hopefully not :/.
> We should worry about our users' security. However, if we don't
> collect survey data, it's harder to improve Freenet. That has
> security implications as well: if Freenet isn't usable, our potential
> users are using less secure alternatives (or not communicating).
> Security risks are worth worrying about, but sufficiently tiny ones
> don't outweigh things that are useful for other reasons, imho.
True, I agree. IMHO it is a fair way to do it as in the Windows
installer: Ask the user kindly, and if he doesn't want to, we shouldn't
force him. Since most of our survey results have disappeared since we
started asking, I take that as a hint that people don't *want to* answer
the surveys - hence I don't think we should try to force them.
- Zero3
___
Devl mailing list
Devl@freenetproject.org
http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Some statistics
Evan Daniel skrev: > On Fri, Aug 7, 2009 at 10:11 AM, Zero3 wrote: > >> True, I agree. IMHO it is a fair way to do it as in the Windows >> installer: Ask the user kindly, and if he doesn't want to, we shouldn't >> force him. Since most of our survey results have disappeared since we >> started asking, I take that as a hint that people don't *want to* answer >> the surveys - hence I don't think we should try to force them. > > I read this differently. If you present the user with "click here to > take our survey" they won't. If you instead present them with "here's > our survey; answer the questions and click here to submit, or here to > not take the survey" you'll get a lot more responses. It's not that > users don't want to take the survey; it's that laziness wins, and if > you stick an extra click in the way you lose most of them. Hmm. You are probably right. The big problem, however, is still that the survey is located on the website... And that we have to launch a web browser to display it. A better (and secure) alternative would be to ask the questions in the uninstaller GUI and publish it to Freenet before uninstalling. But that would require quite some work... - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Some statistics
Evan Daniel skrev: > On Fri, Aug 7, 2009 at 11:06 AM, Zero3 wrote: >> Evan Daniel skrev: >>> On Fri, Aug 7, 2009 at 10:11 AM, Zero3 wrote: >>> >>>> True, I agree. IMHO it is a fair way to do it as in the Windows >>>> installer: Ask the user kindly, and if he doesn't want to, we shouldn't >>>> force him. Since most of our survey results have disappeared since we >>>> started asking, I take that as a hint that people don't *want to* answer >>>> the surveys - hence I don't think we should try to force them. >>> I read this differently. If you present the user with "click here to >>> take our survey" they won't. If you instead present them with "here's >>> our survey; answer the questions and click here to submit, or here to >>> not take the survey" you'll get a lot more responses. It's not that >>> users don't want to take the survey; it's that laziness wins, and if >>> you stick an extra click in the way you lose most of them. >> Hmm. You are probably right. The big problem, however, is still that the >> survey is located on the website... And that we have to launch a web browser >> to display it. >> >> A better (and secure) alternative would be to ask the questions in the >> uninstaller GUI and publish it to Freenet before uninstalling. But that >> would require quite some work... > > If by "some work" you mean "solving the general spam resistance > problem in a way that doesn't involve asking the user to solve > captchas because we have good data to suggest he really doesn't care." The uninstallation survey is just as spamable right now without a CAPTCHA or the like :). Unless of course Google has some fancy anti-spam measures in place. - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Code review, testing, on the beta wininstaller
Matthew Toseland skrev: > Zero3's beta branch of the wininstaller includes a tray applet, allowing the user to conveniently shut down and start up Freenet, browse it and view logs, and doesn't create a new user, which should reduce the amount of trouble we have with antivirus/firewall software. > > In testing (Windows XP, Panda AV): > - The pop-up saying you can use Freenet through the tray icon comes up before the installer finishes, and double clicking at that point doesn't immediately work (but it could just be lag); I double clicked again after it had completed and it brought up Chrome. If the user clicks before the service is started, then you probably ends up with that, yea. I've just changed it so that the tray manager is started *after* the service. Should fix that case. > Other issues not related to Zero3's code: > - Uninstall: If I click yes, the survey comes up, but after entering data, Google fouls up: "Something went wrong. Don't worry about it though. The Spreadsheets Team has been notified and we'll look into it immediately." See bug #3214, maybe we should find an alternative survey provider? This has been broken for quite some time... :< > - Setting the datastore size in the wizard is very slow. It may in fact be preallocating the whole store at that point! We really need to fix this! Bug #3373. > - We need to pick up ?incognito=true in fproxy and change the wizard accordingly - either not show a warning about using a separate browser, or show different text. This is bug #3303. > > Changelog: > - Tray applet! > - Turn off browse after install. > - Install to LocalSystem on win2k, LocalService on XP or later. Do not create the custom user, as it seems to get on badly with anti-virus programs. > - Various fixes and refactoring in installer. > - Update jre_online_installer.exe > - delete unnecessary third party exe's. > - wipe out update.cmd, for now. > - remove unneeded code. > - merge the required disk space and the estimated usage into a single requirement. > - pass in ?incognito=true > - fall back to opening freenet page by just opening the url rather than complaining about no supported browser. > - make uninstaller more robust w.r.t. not running > - better error handling, check for no fproxy.port= > - update many strings. > - various minor changes to start/stop/etc waiting for status logic. > - update src_freenetinstaller/files_install/bin/wrapper-windows-x86-32.exe from 3.3.5, sha1 - update src_freenetinstaller/files_install/lib/wrapper-windows-x86-32.dll > - both these are valid, from http://wrapper.tanukisoftware.org/download/3.3.5/wrapper-windows-x86-32-3.3.5.zip, whose sha256 is dad4277a7981eb8342190df5547804efef76685217b9e125f6d336656f64b066. > - translation updates and placeholders, refactor and comment template. > - script to update translations, put in placeholders, when the main template is altered. Thanks! I hadn't bothered to write such a detailed changelog myself yet. - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] How to start a new translation?
Alex Pyattaev skrev: > I'll start with the client. The installer is used once, the client - forever=) No worries :) - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
[freenet-dev] Updating Windows installs for current user base (specifically: How to install and uninstall the new tray manager?)
I just realized that we have a slight problem with offering to install the tray manager through update.cmd. Users who install the tray manager in this way will have problems uninstalling. Their uninstaller won't be aware of the tray manager, and will therefore not shut it down before trying to delete the installation directory. As the tray manager is executed from within that directory and Windows doesn't allow deletion of a running executable, the uninstaller will throw an "could not delete files" error. The error box will offer to retry, but the user probably won't realize that he needs to manually shut down the tray manager first. Possible solutions: 1) Don't install the tray manager through update.cmd. Users will have to reinstall to get the tray icon. Cons: We are leaving our current user base behind (IMHO: very bad idea) 2) Warn user (upon update.cmd installation) to manually close it down before uninstalling. Cons: The user will probably forget about it and be just as lost when he finally uninstalls. (IMHO: not a proper fix) 3) Update the uninstaller in update.cmd as well. This raises the core issue: That Windows installations soon will have different layout because of the recent change from running the service under a custom user to running under a standardized service user. That gives us 2 possibilities: 3.a) Add backward compatibility to the uninstaller. Cons: Will be a hell to maintain an uninstaller that has to support all previous installation layouts. The recent service user change has resulted in significant changes to it already. (IMHO: an acceptable work-around, but is a PITA to maintain in the long run) 3.b) Update the whole installation in update.cmd. This mainly involves moving current installations away from the custom user and cleaning up after the mess. Cons: Will require some work, and will require either an UAC escalation helper executable for update.cmd or porting update.cmd to real code that can escalate itself. (IMHO: the optimal solution long-term) Anyone? Juiceman, what are your thoughts on this? You are the update.cmd wiz. - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Some statistics
A.Rios skrev: > Hi. I'm not suscribed to the list but I check it sometimes and I want > to share my experience from last week related to the survey. First of > all, the installation fails in my windows 2008 server, saying "Startup > script could not find the Freenet service" (or something like that, it > was a week ago), so I deinstalled from control panel and was brought > to the survey. I filled it and pressed submit button but I get an > error page from Google Docs. I tried a couple of times with no results > and gave up. That's all, maybe you are getting less survey results > because more people are experienced the same problems. Interesting. The Windows installer has a list of OS'es it will install on. Windows Server 2008 is *not* on that list. It probably shares identifier with Vista, as they share the same codebase. Regarding the error, we suspect that it might be related to the custom service user we currently create and use to run Freenet. If you have the time, please try the latest beta of the installer and see if the problem still occurs: Binary: http://privat.zero3.dk/FreenetInstaller_Beta.exe Source: http://github.com/freenet/wininstaller-staging/tree/beta You can reply directly to my mail address if you prefer not subscribing to the developer mailing list. - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Some statistics
A.Rios skrev: > On Sun, Aug 16, 2009 at 4:01 PM, Zero3 wrote: >> Interesting. The Windows installer has a list of OS'es it will install on. >> Windows Server 2008 is *not* on that list. It probably shares identifier >> with Vista, as they share the same codebase. >> >> Regarding the error, we suspect that it might be related to the custom >> service user we currently create and use to run Freenet. If you have the >> time, please try the latest beta of the installer and see if the problem >> still occurs: >> >> Binary: http://privat.zero3.dk/FreenetInstaller_Beta.exe >> Source: http://github.com/freenet/wininstaller-staging/tree/beta >> >> You can reply directly to my mail address if you prefer not subscribing to >> the developer mailing list. >> > > Nice, it installed without a warning and was able to browse freenet > within minutes :) It just complained about the build being "too old" > and some plugins that couldn't be loaded, but I assume is just a "beta > feature" ;) Cool! Yea, the included version of Freenet and the plugins might be slightly outdated, as the beta installer is only updated once in a while when I implement new stuff. Freenet should update itself automatically. Plugin autoupdating is on the way too :). Note that the beta installer doesn't have a working update.cmd yet (one of the few remaining things we need to fix before releasing this on the website). - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] More wininstaller woes
Matthew Toseland skrev: > When can we deploy the new wininstaller? > > [02:19:53] help me guys > [02:20:00] i installed freenet > [02:20:29] but it says Freenet starter was unable to control the > Freenet system service > [02:20:34] what do i do? > [02:21:21] videomax: windows? > [02:21:25] yes > [02:21:28] xp or vista? > [02:21:29] windows xp > [02:21:32] 32bit > [02:23:27] using 1230 installer > [02:23:52] Reason: Service did not respond to signal > [02:24:53] i cant start freenet manually with net start either > [02:25:24] check c:\program files\freenet\wrapper.log ... any error > in that file? > [02:25:48] no errors > [02:25:53] STATUS | wrapper | 2009/08/17 21:23:32 | Freenet > background service installed. > [02:26:48] just one line, nothing more? > [02:26:54] nothing more > [02:28:12] I just tried installing it in C:\freenet > [02:28:22] like maybe the spaces in Program Files messed it up > [02:28:26] didnt work though > [02:30:47] im gonna try a reboot > [02:30:52] be back soon > [02:36:39] no such luck yo > [02:36:51] i am also running tor > [02:36:58] videomax: tor shouldn't matter. > [02:36:58] videomax: do you have java installed? > [02:37:00] do you think that is causing the problem? > [02:37:05] yes i have java > [02:37:19] videomax: What version of Java by what vendor? > [02:37:57] videomax: Also, do you have restrictive password / > account creation policies in place, and are you running any antivirus / > firewall apps? > [02:38:17] jre 1.6.0_07 > [02:38:20] from sun > [02:38:32] not running any antvirus/firewall apps > [02:38:37] i am admin > [02:39:24] videomax: That should work. However, there are is a > remote code execution vulnerability in that version of Java; you should > upgrade *immediately*. > [02:39:47] whoah > [02:39:53] right when you said that > [02:40:00] the little update java box opned > [02:40:02] videomax: uninstall it and install a newer version > [02:40:14] don't update, just uninstall and reinstall > [02:40:26] To be specific, you need 1.6.0_15 or later, or 1.5.0_20 > or later. > [02:40:36] hmm i means don't have the "upgrade" function... > [02:40:43] and then freenet will work? > [02:40:43] uninstall it and install a newer one > [02:40:56] it should work. > [02:42:33] I wouldn't think Freenet would break because of that Java > version; it's not that ancient. But it's worth trying, and you *definitely* > want to fix it regardless. > [02:43:52] k > [02:44:15] hmm it is not the java version, but the java installation > [02:44:26] that's why i said uninstall/reinstall > [02:46:05] sdiz: Weird. I'm glad I don't have Windows :/ > [02:46:39] sdiz: If that's the issue, how hard would it be to > generate a log message to the effect of "Your Java is broken; fix it." > [02:52:44] ok > [02:52:51] new version of java installed > [02:52:56] other is uninstalled > [02:53:09] should i reinstall freenet? > [02:54:25] ok > [02:54:29] reinstalled freenet > [02:54:33] still getting same error > [02:56:16] is there anything else i can do?? > [02:59:44] FATAL | wrapper | 2009/08/17 21:59:20 | Unable to > access registry to obtain environment variables - The operation completed > successfully. (0x0) > [02:59:55] im getting this error in wrapper.log now > [03:03:51] ah well > [03:03:56] i guess it waasnt meant to be > [03:24:19] Sigh, he left :/ > [03:24:35] My internet was acting up or I would have helped out... > [03:25:46] Sounds like a (relatively) simple permissions issue. > [03:25:56] Yeah. > [03:26:03] Notably, *not* the same issue he had before. Never heard of the wrapper error before, but nevertheless out of my hands. He really should try the beta though, as both errors might be related to the custom user we create atm. The beta is generally much more sane in terms of system permissions. The main issue blocking a release is policy of updating current installations. I could use your comments on that thread. Atm. I lean towards porting update.cmd to real code (Juiceman suggested that he might learn AHK (in which case I can give a hand of course)). (I've just moved halfway across the country, so I will be busy moving in for the next week or so and not have any time for Freenet. I'll keep participating in the discussions on this list though.) - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
[freenet-dev] Translators wanted for the next version of the Windows Installer!
(To dev list, bcc'ed to existing translators of which I have e-mail addresses) Hi all The next version of the wininstaller is more or less done. This version covers many changes, including changes to localizable text strings. To make the translation process less repetitive, many strings have been split into substrings that are reused throughout the installer. This means more strings, but much shorter ones (and a quicker job in the end!). Overlapping strings from the existing translations - Danish, German, Finnish, French and Italian - have been reused. If you are the previous translator for one of those languages and do not wish to continue the job, please tell me so I can look out for someone else to do it :). It would be *totally* sweet to have more translations in this version, so if you can translate from English into another language than the above mentioned, please consider doing so! Existing translators: Simply update your new translation file located in the beta branch of wininstaller-staging (http://github.com/freenet/wininstaller-staging/tree/beta/src_translationhelper). New translators: Please read the quick guide inside the translation template (http://github.com/freenet/wininstaller-staging/blob/beta/src_translationhelper/TranslationTemplate.inc) (existing translators might want to read the updates here as well) Submit your translation to wininstaller-staging, this mailing list or directly to my mail address. Thanks! - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Testing on Windows
Matthew Toseland skrev: > tried 1231 installer on a windows system: > > freenet installer pops up before java installer has finished. is this a > problem? It shouldn't be a problem. The wininstaller waits for the JRE online installer to return before popping back up. Unfortunately, it returns before it is actually done installing the JRE (the rest is apparently done in a forked process or something). Poor coding on Sun's part... As an extra precaution (besides checking the registry for correct JRE version info, and waiting for the JRE installer to return), the installer will not pop back up before the main runtime library exists in the JRE install folder. I've done a race condition test on this without any apparent problems. - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Testing on Windows
Matthew Toseland skrev:
> Here's another one:
>
> I accidentally built the installer with the beta branch. The tray icon
> worked, but when I uninstalled, and told it to do the survey, the survey
> failed (as usual), but control panel hung. Ideas? I think it might be waiting
> for firefox to close, but this is very bad behaviour, as there may be other
> stuff in firefox?
>
> I don't think this is specific to the beta branch - we wait for firefox (or
> whatever browser) to close before closing control panel.
Nop, it will not wait for the browser to close. It will exit right after
launching:
[CODE]
If (_DoSurvey)
{
Run, http://freenetproject.org/uninstall.html, , UseErrorLevel
}
Exit()
[/CODE]
(The execute-and-wait-for-it-to-finish command is called "RunWait" as
opposed to the "Run" used here which will continue right away)
Are you sure that freenetuninstaller.exe is running while the control
panel hangs?
- Zero3
___
Devl mailing list
Devl@freenetproject.org
http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] What's this translation file?
Doh, that translation was never ported from master to beta. The file lacks the credit header though. If you can supply me with a header like the following example, I'll port the translation to the beta branch at once. Translation file header example: ; ; Translation file - French (fr) - by Romain Dalmaso (artefact2 that-a-thingy gmail that-dot-thingy com) ; - Zero3 Caco Patane skrev: > Those are the spanish translations for the wininstaller, at the time > commited they were working fine and tested (correct display of the > strings without messing up the interface) in a windows VM. Don't know > if they continue to be compatible because I didn't followed > wininstaller development. > > Maybe I can take a look on the weekend... any significant change in > the wininstaller regarding language? > > On Fri, Aug 21, 2009 at 5:32 PM, Matthew > Toseland wrote: >> Found this in my working copy, but I was on beta for some reason so it's >> confusing... What should be done with it? Is it compatible with the current >> master or beta for the wininstaller? > ___ > Devl mailing list > Devl@freenetproject.org > http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Testing on Windows
Matthew Toseland skrev:
> On Friday 21 August 2009 20:12:39 Zero3 wrote:
>> Matthew Toseland skrev:
>>> Here's another one:
>>>
>>> I accidentally built the installer with the beta branch. The tray icon
>>> worked, but when I uninstalled, and told it to do the survey, the survey
>>> failed (as usual), but control panel hung. Ideas? I think it might be
>>> waiting for firefox to close, but this is very bad behaviour, as there may
>>> be other stuff in firefox?
>>>
>>> I don't think this is specific to the beta branch - we wait for firefox (or
>>> whatever browser) to close before closing control panel.
>> Nop, it will not wait for the browser to close. It will exit right after
>> launching:
>>
>> [CODE]
>> If (_DoSurvey)
>> {
>> Run, http://freenetproject.org/uninstall.html, , UseErrorLevel
>> }
>>
>> Exit()
>> [/CODE]
>>
>> (The execute-and-wait-for-it-to-finish command is called "RunWait" as
>> opposed to the "Run" used here which will continue right away)
>>
>> Are you sure that freenetuninstaller.exe is running while the control
>> panel hangs?
>
> I don't know. I do know that control panel hangs until the browser exists.
Odd. Maybe it also considers processes spawned by the uninstaller as
part of the uninstaller itself, and does not return control to the
control panel until all of these have terminated. Would make sense, as
uninstallers often continue in other processes than the originally
executed one (the wininstaller uninstaller included).
Nevertheless, this is a design choice by Microsoft. If they want to
freeze out the user while any part of an uninstaller is running, I
shouldn't try to (and probably can't) get around it.
A possible workaround could be to add a message to the survey completion
page simply asking the user to close the window.
- Zero3
___
Devl mailing list
Devl@freenetproject.org
http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Updating Windows installs for current user base (specifically: How to install and uninstall the new tray manager?)
Juiceman skrev: > On Wed, Aug 26, 2009 at 10:54 AM, Matthew > Toseland wrote: >> On Wednesday 26 August 2009 01:27:18 Juiceman wrote: >>> On Tue, Aug 25, 2009 at 7:57 PM, Matthew >>> Toseland wrote: >>>> On Saturday 22 August 2009 06:33:32 Juiceman wrote: >>>>> On Tue, Aug 11, 2009 at 10:21 PM, Juiceman wrote: >>>>>> On Sun, Aug 9, 2009 at 8:43 PM, Zero3 wrote: >>>>>>> I just realized that we have a slight problem with offering to install >>>>>>> the tray manager through update.cmd. >>>>>>> >>>>>>> Users who install the tray manager in this way will have problems >>>>>>> uninstalling. Their uninstaller won't be aware of the tray manager, and >>>>>>> will therefore not shut it down before trying to delete the installation >>>>>>> directory. As the tray manager is executed from within that directory >>>>>>> and Windows doesn't allow deletion of a running executable, the >>>>>>> uninstaller will throw an "could not delete files" error. >>>>>>> >>>>>>> The error box will offer to retry, but the user probably won't realize >>>>>>> that he needs to manually shut down the tray manager first. >>>>>>> >>>>>>> Possible solutions: >>>>>>> >>>>>>> 1) Don't install the tray manager through update.cmd. Users will have to >>>>>>> reinstall to get the tray icon. Cons: We are leaving our current user >>>>>>> base behind (IMHO: very bad idea) >>>>>>> >>>>>>> 2) Warn user (upon update.cmd installation) to manually close it down >>>>>>> before uninstalling. Cons: The user will probably forget about it and be >>>>>>> just as lost when he finally uninstalls. (IMHO: not a proper fix) >>>>>>> >>>>>>> 3) Update the uninstaller in update.cmd as well. This raises the core >>>>>>> issue: That Windows installations soon will have different layout >>>>>>> because of the recent change from running the service under a custom >>>>>>> user to running under a standardized service user. That gives us 2 >>>>>>> possibilities: >>>>>>> >>>>>>> 3.a) Add backward compatibility to the uninstaller. Cons: Will be a hell >>>>>>> to maintain an uninstaller that has to support all previous installation >>>>>>> layouts. The recent service user change has resulted in significant >>>>>>> changes to it already. (IMHO: an acceptable work-around, but is a PITA >>>>>>> to maintain in the long run) >>>>>>> >>>>>>> 3.b) Update the whole installation in update.cmd. This mainly involves >>>>>>> moving current installations away from the custom user and cleaning up >>>>>>> after the mess. Cons: Will require some work, and will require either an >>>>>>> UAC escalation helper executable for update.cmd or porting update.cmd to >>>>>>> real code that can escalate itself. (IMHO: the optimal solution >>>>>>> long-term) >>>>>>> >>>>>>> Anyone? >>>>>>> >>>>>>> Juiceman, what are your thoughts on this? You are the update.cmd wiz. >>>>>>> >>>>>>> - Zero3 >>>>>> There is no easy answer. We don't want to leave users behind but we >>>>>> can't maintain backwards compatibility. >>>>>> 3a) and I can have update.cmd download a version for these older >>>>>> installs. >>>>>> >>>>>> As far as migrating older installs, UAC does present problems. I >>>>>> guess if you could make an UAC escalation helper to boost update.cmd >>>>>> that could work. >>>>>> >>>>>> Regarding porting update.cmd to real code, I could try to learn AHK >>>>> I have started looking into AHK, it seems fairly easy. I already have >>>>> a UAC escalation helper figured out. >>>> What is the status of this? I understand it is the main thing preventing >>>> us from releasing the new installer? Have you decided to go with solution >>>> 3b? >>> The sticking point is with old i
Re: [freenet-dev] Detecting downloads dir in the installer?
Matthew Toseland skrev: > When the installer runs, even if it is elevated, it is running as the > installing user, correct? Therefore, we could detect that user's download dir > (My Downloads etc)? > > See this bug: > https://bugs.freenetproject.org/view.php?id=3343 > > Thoughts? You are correct. I've commented on the bug :). - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] freenet/wininstaller-staging.git
Hi Thank you very much! I'm very busy at the moment, but I will test and commit this as soon as I get the time! - Zero3 (maintainer of the Windows installer) Валентин StandAlone-Alien skrev: > Russian translation > > > > > ___ > Devl mailing list > Devl@freenetproject.org > http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Wininstaller it_IT l10n update 090901
Luke771 skrev: > attached. > includes all the new strings that were added ... months ago :/ > also some minor rewording > I was totally sure that I had sent this in long ago. sorry > please commit Thanks! No worries about the delay. I'm overbooked myself atm., so it will take a little while before I get this in git too. - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Freenet-Cards: ID cards with your node-ref
Robert Hailey skrev: > I really like the *idea*, but I wish it were easier to implement. It > would be a pain to enter that huge noderef by hand. I'd love to see this too. We have a couple of long-standing reports about this already though: https://bugs.freenetproject.org/view.php?id=166 (Proper invite mechanism ) https://bugs.freenetproject.org/view.php?id=1343 (Darknet: short references / out of band password verification) https://bugs.freenetproject.org/view.php?id=1872 (Short noderefs) - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Thoughts on HTL data
Hey Just a short note from me: Kudos to you for from here for digging into all this technical and statistical stuff (of which I don't understand the half of :P). It's nice to see some work being done on improving Freenet at this level too! :). - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
[freenet-dev] [wininstaller beta] update.cmd wininstaller beta compatibility
Hey What is the status for wininstaller beta compatibility in update.cmd Juiceman? I've noticed the following bugs while testing today: 1) Seems like it tries to set file permissions on installs without the custom user: "- Checking file permissions - Changing file permissions freenet: Der blev ikke udført nogen afbildning mellem kontonavne og sikkerheds-id." A work-around until we have a real updater would be to not do so if bin\freenettray.exe exists. 2) The tray manager is not being updated. (I agree that we shouldn't install it on old installs for now, but if it already exists, it should be updated as all the other executables) - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
[freenet-dev] [wininstaller beta] Work on implementing the Russian translation
Hi again Sorry for the long delay. I've looked at your translation today, but it seems like the installer might not support Cyrillic characters as it is right now. They show as "" on my system at least. Did you test the translation yourself by compiling the installer with your translation? Did it work? I've filed a bug for it at https://bugs.freenetproject.org/view.php?id=3563 if you want to follow the progress on this. I've got some things to try out when I have the time. - Zero3 Валентин StandAlone-Alien wrote: > Russian translation > > > > > ___ > Devl mailing list > Devl@freenetproject.org > http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
[freenet-dev] [wininstaller beta] Release candidate
Hi all Time for a proper release candidate of the wininstaller beta branch. As always, all kinds of constructive feedback is welcome. Source: http://github.com/freenet/wininstaller-staging/tree/beta Binary: http://privat.zero3.dk/FreenetInstaller_Beta.exe Most interesting changes in the release candidate are: * Made launcher silently fail if start.exe failed (instead of trying to launch anyway when service clearly isn't running). * Updated Java online installer to version 1.6u16. * Bumped required Java version to 1.6. 1.5 is going end-of-life very soon. * Added Spanish translation. * Updated Danish translation. * Added update.cmd (still not perfect, see separate mail) * Added a wrapper.restart.delay to wrapper.conf as the wrapper default might be 0 seconds according to some reports. The delay might be needed in rare cases. * Delay execution of the tray manager until the user has clicked through the "Install done" infobox. (Usability reasons) * Rework of disk space calculation and display. Unless anything serious shows up, this should be ready to replace the master branch :). - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
[freenet-dev] [wininstaller beta] Testing on Windows + Uninstall survey
Matthew Toseland wrote:
> On Monday 24 August 2009 17:03:21 Zero3 wrote:
>> Matthew Toseland skrev:
>>> On Friday 21 August 2009 20:12:39 Zero3 wrote:
>>>> Matthew Toseland skrev:
>>>>> Here's another one:
>>>>>
>>>>> I accidentally built the installer with the beta branch. The tray icon
>>>>> worked, but when I uninstalled, and told it to do the survey, the survey
>>>>> failed (as usual), but control panel hung. Ideas? I think it might be
>>>>> waiting for firefox to close, but this is very bad behaviour, as there
>>>>> may be other stuff in firefox?
>>>>>
>>>>> I don't think this is specific to the beta branch - we wait for firefox
>>>>> (or whatever browser) to close before closing control panel.
>>>> Nop, it will not wait for the browser to close. It will exit right after
>>>> launching:
>>>>
>>>> [CODE]
>>>> If (_DoSurvey)
>>>> {
>>>>Run, http://freenetproject.org/uninstall.html, , UseErrorLevel
>>>> }
>>>>
>>>> Exit()
>>>> [/CODE]
>>>>
>>>> (The execute-and-wait-for-it-to-finish command is called "RunWait" as
>>>> opposed to the "Run" used here which will continue right away)
>>>>
>>>> Are you sure that freenetuninstaller.exe is running while the control
>>>> panel hangs?
>>> I don't know. I do know that control panel hangs until the browser exists.
>> Odd. Maybe it also considers processes spawned by the uninstaller as
>> part of the uninstaller itself, and does not return control to the
>> control panel until all of these have terminated. Would make sense, as
>> uninstallers often continue in other processes than the originally
>> executed one (the wininstaller uninstaller included).
>>
>> Nevertheless, this is a design choice by Microsoft. If they want to
>> freeze out the user while any part of an uninstaller is running, I
>> shouldn't try to (and probably can't) get around it.
>
> There must be a way to detach it.
There might be. No idea how much hacking it would take though. IMHO I
don't think the minor usability issue is worth the hack.
On Vista I am able to close the control panel, although not start a new
uninstall before the browser is closed.
>> A possible workaround could be to add a message to the survey completion
>> page simply asking the user to close the window.
>
> Unfortunately it's broken atm.
Any update on this? Please at least remove the survey from
http://freenetproject.org/uninstall.html and replace it with a
"Temporary out of order" message or something. We are seriously wasting
people's time right now.
- Zero3
___
Devl mailing list
Devl@freenetproject.org
http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] [freenet-support] WinXP installer fails, service won't start
Hi I'm the maintainer of the Windows Installer. Sorry for the long delay of this reply - I've been extremely busy lately. If you can, please send us the wrapper.log logfile. You can find it via the tray icon in the beta release. It might contain useful information. Where do you get the "2" error code? From the Windows service manager? Is there anything special about your system? Group policies? Special version of Windows? Non-standard antivirus software? You should be able to run the service in a command promt by simply copying the command line from the Windows service manager, eventually removing the "-s" switch. - Zero3 Magnus Ekhall wrote: > Matthew Toseland skrev: >> On Tuesday 25 August 2009 18:34:34 Magnus Ekhall wrote: >>> I tried to install using the 1232 version of the XP installer. >>> >>> At the end of the installation it says that it failed becaus the service >>> could not be started. >>> >>> If I try to start the service manually it will eventually fail as well >>> with the error code "2". >>> >>> Any ideas? >> Try the beta installer: >> http://privat.zero3.dk/FreenetInstaller_Beta.exe >> >> This will install an old version of Freenet, with no update.cmd. But it >> should auto-update to the latest version in an hour or so. >> >> Uninstall your current Freenet first, but don't uninstall the Java version >> it installed. >> >> PS Zero3: Shall I compile up a more recent beta/ version? > > > I tried the beta installer, but it too could not start the freenet service. > > "Service did not respond to signal" it says. > > I'm runnung the installer as admin and I have a recent java. > > Can I run the service in a shell or something to get a bit better error > messages? > > ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Win64 support options?
xor wrote: > On Friday 21 August 2009 22:31:03 Matthew Toseland wrote: >> How common is 64-bit Vista? Currently we install a 32-bit JVM, which works, >> but doesn't get auto-updated, and is somewhat slower than if we'd installed >> a 64-bit one (assuming we fix the FEC libraries and manage to build them >> for Windows). We can't install a 64-bit JVM, because the free version of >> the Java Service Wrapper (which we use for self-restarting the node when >> deploying updates, and for detecting hangs) only supports 32-bit. >> >> Options? > > Vista64 is very common already and Win7 x64 will be even more widespread. 64 > bit CPUs have been there for ages and we should support them. > > Regarding the 64bit JVM I have to state that we should keep shipping the > 32bit > one - it still seems to be intended by Sun that 32bit java is used on 64bit > windows: > > - The 64bit JVM does not install a Firefox plugin, or at least it does not > work (maybe because Firefox is still 32bit?) > > - The auto updater of the 64 bit JVM installs a 32 bit JVM so then you have > 2. > This is obviously a bug but it shows that Sun does not properly maintain the > 64bit windows JVM yet. I'm a bit out of the loop on this issue. Did we figure out if we should keep installing 32-bit java on 64-bit Windows? If so, any idea of when it will change? - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] [wininstaller beta] update.cmd wininstaller beta compatibility
Juiceman wrote: > On Fri, Oct 2, 2009 at 7:31 PM, Zero3 wrote: >> Hey >> >> What is the status for wininstaller beta compatibility in update.cmd >> Juiceman? > > I'm sorry, I've not progressed very far on the AHK replacement, I got > distracted. I don't know when I will be able to work on it again. I > will commit the file I started in case someone else wants to get to > it. No worries. I didn't mean the new updater though, but rather if the current update.cmd completely supports the beta branch (tray manager etc.)? >> I've noticed the following bugs while testing today: >> >> 1) Seems like it tries to set file permissions on installs without the >> custom user: >> >> "- Checking file permissions >> - Changing file permissions >> freenet: Der blev ikke udført nogen afbildning mellem kontonavne og >> sikkerheds-id." >> >> A work-around until we have a real updater would be to not do so if >> bin\freenettray.exe exists. > > I will commit this asap. Cheers :) >> 2) The tray manager is not being updated. (I agree that we shouldn't >> install it on old installs for now, but if it already exists, it should >> be updated as all the other executables) >> >> - Zero3 > > As far as I know, the tray manager is not hosted on the Freenet > website anywhere and therefore cannot be updated. toad, can you upload a dummy of this, until the beta has been deployed? - Zero ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] [wininstaller beta] update.cmd wininstaller beta compatibility
Matthew Toseland wrote: > On Thursday 08 October 2009 17:30:45 Zero3 wrote: >> Juiceman wrote: >>> As far as I know, the tray manager is not hosted on the Freenet >>> website anywhere and therefore cannot be updated. >> toad, can you upload a dummy of this, until the beta has been deployed? > > Done. Sorry for the delay. It would be really great to get the new installer > deployed, it is IMHO vital for 0.8. As I understand it the big problem is > that there is no working update script for the new installer? The beta branch is basically ready to be merged into master and deployed. This will also finally resolve the last long-standing wininstaller bugs from back when we used the Java installer (the custom user related ones). Only remaining issue is indeed update.cmd, but Juiceman hasn't replied on whether development for the beta branch is finished. It seemed to work in my tests though. Of the reported wininstaller failures, the beta seems to solve most of them. A few have reported failure even with the beta. I only managed to get a hold of one of these reporters, and his problem was revealed to be a broken Windows registry database. This new install design (using a standard service user account instead of a custom one) raised a new issue regarding how to update old installs. It was more or less agreed that it would be a better idea to start from scratch on a real updater than hacking UAC-functionality into update.cmd. Juiceman has looked at this, but not made any real progress. I'm insanely overbooked at the moment, so I'm afraid I cannot do this myself right now. This means that we will be leaving the current users behind, most importantly leaving them without the upcoming tray manager (which IMHO is an important aspect both short- and long-term). - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] [wininstaller beta] update.cmd wininstaller beta compatibility
Juiceman wrote: > Commit > http://github.com/freenet/wininstaller-staging/commit/2fe991d7c58f3fe2c23b222ee678ea5312c87072 > should have fixed update.cmd compatibility, please deploy it to the > website. > [snip] > Once the freenettray.exe is on the website I can wire it in to at > least update installs going forward. No need to wait on that. Deploy > away. > > Upgrading old installs can come later. > Cool! toad you should look into deploying the beta then. I have no idea how to handle this in git though. Perhaps it is possible to rename the "master" branch to "deprecated_v1" and the "beta" branch to "master"? Or would it be better to simply merge the branches and tag the merge with "generation_2" or something? - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] test install on Windows 7
Wee! But. Was that with the current version (the one from the website) or with the beta? I'm very surprised if the current version works on Win7 :o. Can't complain though... - Zero3 On Fri, 23 Oct 2009 11:30:33 -0500, Ian Clarke wrote: > I just did a fresh install of Freenet on a newly-minted Windows 7 box, > and I'm pleased to say that it went very smoothly! > > The installer downloaded and installed Java, and while (of course) I'm > no newbie, I didn't really encounter any usability problems, except > perhaps all the reading a user is expected to do when you choose > security settings. > > I think with those options we should try reducing it to a single > sentence for each, with a "read more..." link which opens up a more > detailed explanation. > > Ian. ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] New wininstaller
Awesomeness! Did the merge succeed without issues? Any problematic conflicts? The reason for the "Browse Freenet" to "Launch Freenet" rename in the first place, was that Freenet is starting to do much other stuff than "browsing [websites]". Mail, forums, IM, file sharing. "Browse" sounds a bit misleading as a common verb for that. Maybe something completely different? I'm responsible for turning the incognito flag back on. I really think the block should be placed in Freenet (by simply checking the user-agent), as the block can then easily be removed on a new build when Google have fixed Chrome. If it's placed in the launcher, we can't push the update to enable it again later on, as we depend on people updating their helper executables themselves. Which they probably won't (it's possible via the new tray manager though). - Zero3 On Fri, 23 Oct 2009 20:56:23 +0100, Matthew Toseland wrote: > I have merged the beta branch of the wininstaller, in accordance with > instructions from Juiceman and Zero3. The main changes in this branch are > that it doesn't create a custom user to run Freenet in (which caused many > problems for many users installing it due to password policies, anti-virus > software, and things we don't understand), and that it has a system tray > icon, from where the user can conveniently (provided his system tray isn't > hopelessly cluttered already) start or stop or browse Freenet. > > In testing, it works, and so does Freenet. However there are a number of > issues we need to deal with: > - "Launch Freenet after installation" - Freenet will run anyway, launching > it means browsing it, maybe we should change "Launch Freenet" here and in > the system tray menu to "Browse Freenet"? > - TUFI menus don't show up well in Chrome. Chrome is our default browser on > Windows because it has incognito mode, which we use. The author of TUFI > should fix the stylesheet so that the drop-down menus are usable on Chrome, > please! (Can somebody please forward this request via the FMS board > please?) > - Chrome sometimes seems to lose the web interface CSS, showing either a > blank screen or a non-styled copy of the loading a page or dangerous > content page. There were no errors in the log concerning fproxy. Reload and > even shift+reload make no difference, but restarting Chrome fixes this. > Initially I had thought it might be related to it using lots of connections > as I was loading many sites with many images, but that didn't seem to make > any difference. Thoughts? > - We have a flag to tell Freenet that we are running Chrome in incognito > mode, which will show a much shorter and less worrying browsers warning at > the beginning of the wizard. However, bugs in Chrome mean that just because > we tell Chrome to open Freenet in incognito mode doesn't mean it actually > will. I had turned off this flag, Juiceman turned it back on, for now I > have turned it back off, because it is a security issue. Thoughts? > > The next build will be released with the new installer in any case. ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Improving text on wizard was Re: test install on Windows 7
On Fri, 23 Oct 2009 23:51:56 +0100, Matthew Toseland wrote: > WELCOME SCREEN: IMHO: Remove it. If needed, add a "Configure manually" button in the corner of the rest of the wizard pages. 1 click saved. > If we are browsing in incognito mode, we have a shorter warning: IMHO: Don't bother the user when he is doing the right thing. He is safe, so let him in without questions. 1 click saved. > Unfortunately starting Chrome with the incognito flag does not reliably > ensure the window is opened in incognito mode - if Chrome is already > running, it will open it in a non-incognito window/tab. So at the moment > this is turned off. So afaics we are waiting for Google to fix it? Firefox > is likely to have similar issues based on my experience with profiles, > although it may be possible to work around that with -no-remote. Does FF3.5 > have an equivalent of incognito mode? We are indeed waiting for Google to fix it. I'm keeping an eye on their bug tracker. Regarding FF, we are awaiting a command line option to use private browsing which will be implemented in 3.6 (https://developer.mozilla.org/en/Command_Line_Options#-private). > AUTO UPDATE AND PLUGINS IMHO: Default to autoupdate without asking (on Windows and Mac - should be off on Linux when installed from a package). Nodes will quickly be locked out of the network if they don't update, rendering their nodes useless. 1 click saved. > WELCOME: IMHO: Remove this page and show it as some kind of status message on the fproxy main page. 1 click saved. - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] New wininstaller
On Sat, 24 Oct 2009 00:56:49 +0100, Matthew Toseland wrote: > On Friday 23 October 2009 23:54:36 Matthew Toseland wrote: >> On Friday 23 October 2009 23:27:34 zero3 wrote: >> > >> > Awesomeness! >> > >> > Did the merge succeed without issues? Any problematic conflicts? >> >> Several files had issues, I sided with the beta branch. >> > >> > The reason for the "Browse Freenet" to "Launch Freenet" rename in the >> > first >> > place, was that Freenet is starting to do much other stuff than >> > "browsing >> > [websites]". Mail, forums, IM, file sharing. "Browse" sounds a bit >> > misleading as a common verb for that. Maybe something completely >> > different? >> >> Open Freenet? >> > >> > I'm responsible for turning the incognito flag back on. I really think >> > the >> > block should be placed in Freenet (by simply checking the user-agent), >> > as >> > the block can then easily be removed on a new build when Google have >> > fixed >> > Chrome. If it's placed in the launcher, we can't push the update to >> > enable >> > it again later on, as we depend on people updating their helper >> > executables >> > themselves. Which they probably won't (it's possible via the new tray >> > manager though). >> >> Hmmm, okay. It is disabled in fproxy at the moment, so I will reinstate >> the change you made, but I will add comments on both sides. >> > Done. Any thoughts on Launch Freenet vs Open Freenet? We do tell the user > that it will run in the background anyway, so maybe launch is okay? Roger that. Either works for me. I don't really have an opinion on which is best. The Danish translation for either of them would be the same, actually. - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] test install on Windows 7
On Fri, 23 Oct 2009 18:25:17 -0500, Ian Clarke wrote: > On Fri, Oct 23, 2009 at 5:10 PM, zero3 wrote: >> But. Was that with the current version (the one from the website) or with >> the beta? I'm very surprised if the current version works on Win7 :o. >> Can't >> complain though... > > It was with the one currently on the website, and it appeared to work > exactly as intended so far as I can tell. :o. I can't complain though :) - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Improving text on wizard was Re: test install on Windows 7
On Sat, 24 Oct 2009 20:19:34 +0100, Matthew Toseland wrote: >> > If we are browsing in incognito mode, we have a shorter warning: >> >> IMHO: Don't bother the user when he is doing the right thing. He is safe, >> so let him in without questions. 1 click saved. > > Problem is it is hard to impossible to reliably detect when he is doing the > *wrong* thing. Until we have a custom browser with non-localhost url's, we > need the user to be aware of the issue? If the incognito flag is passed, I think we can safely assume that we are actually in incognito mode and not show any message. With the exception of the current Chrome bug of course. When Chrome is fixed and FF 1.6 is out, we can cover quite some ground. We still have the separate IE warning, don't we? - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
[freenet-dev] Depency on emu
Speaking of servers, how is our depency on emu looking? There was some discussion a while ago to move things off emu to save the monthly expense? - Zero3 Matthew Toseland wrote: > On Tuesday 20 October 2009 04:46:10 Ian Clarke wrote: >> I reported this bug a few days ago: >> >> https://bugs.freenetproject.org/view.php?id=3604 >> >> Read it for the details, but basically I'm seeing an intermittent >> problem where Webstart can't download the installer. Is anyone else >> seeing it? > > Reproduced the bug, fixed it by pointing the JWS installer to Google Code. > Which means that we now only use the mirror network for update.sh / > update.cmd. The freenet.jnlp itself is served from emu directly, so the JNLP > still says it's downloading from checksums.freenetproject.org. > > > > > ___ > Devl mailing list > Devl@freenetproject.org > http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Update on getting rid of emu
Matthew Toseland wrote: > We need to migrate the Wikka wiki to MediaWiki, because the latter is > standard and we will be able to host it externally. E.g. sourceforge Hosted > Apps allows for data import for MediaWiki. The wiki isn't really kept enough up-to-date, is it? I personally think the idea of something more integrated (like Trac) would be awesome. MediaWiki seems a bit over-do for Freenet? > BUG TRACKER: > - Dev intelligence i.e. stuff people have said. If these are corroborated > quickly they should be acted upon, else they should be closed. Something integrated (same user account, for starters) might help motivating people to put this stuff in the wiki as well. > If we keep the bug tracker: > - We need to find somewhere to host MANTIS. Probably we will have to pay for > this. > - We need to keep it up to date ourselves, which is somewhat involved. It may > not be as bad as nextgens implies though. > - Minimum immediate work. I personally think Mantis is *very* bad usability-wise. Trac, Launchpad, and many other bug trackers are much easier to use. If we even have to pay just to keep that thing running, I'd say find something else. > If we don't keep the bug tracker: > - We can use any hosted bug tracker anywhere. E.g. Sourceforge Hosted Apps > includes both Mantis and Trac. We will likely be able to avoid any fixed > monthly payments. > - We can use any bug tracker: Mantis, Trac, etc. See below. > - We will need to do a "spring clean": Keep the current bug tracker up for a > while but read-only, *manually* migrate any important bugs and issues to the > new tracker. > - This will be significant work. > - It will involve going over the bugs, dumping those which are out of date, > abandoned etc, and rewriting those bugs and feature issues that are still > valid. Trac's wiki functionality may be useful for this, although it loses > the ability to link bugs formally. > - It may be a useful exercise in terms of prioritising and de-junking. > > However, we have 20 weeks left of funding, so we have to ask whether spending > a week de-junking is worth it? If it comes down to costing us £40-£80 per month... It quickly runs up. I'm up for giving a hand and doing 5 'a day of the de-junking and moving. I'm sure there are more people around willing to give a hand. > An important related point: Relatively few end-users use the current bug > tracker. It is on the Contribute menu on the website, but the main reason > IMHO is it is not very newbie friendly. Uservoice is a reasonable solution > for end-user feature suggestions and gauging public opinion, but because it > does not force users to register their email addresses, it is worthless for > solving individual reproducible bugs. It might reasonably be argued that we > should have a separate issue tracker or forums system for end-user bug > reports. Also, it may make sense for the developer-oriented bug tracker to be > open source, whereas it matters less for the end-user tracker, because 1) > end-users care less, and 2) long term stuff with detailed implementation > notes is likely to be on the developer-oriented bug tracker. Again, let's find something more user-friendly than Mantis :/ > > If this line of reasoning is correct, we need to choose an end-user-oriented > issue tracker or forums system (either way ideally gratis and hosted) to > complement Uservoice. Suggestions? It would make sense to find a tracker that both users and devs can use. Saves the overhead of moving things from e.g. a forums system to a bug tracker. - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Update on getting rid of emu
Matthew Toseland wrote: > On Friday 30 October 2009 17:10:02 Zero3 wrote: >> Matthew Toseland wrote: >>> If this line of reasoning is correct, we need to choose an >>> end-user-oriented issue tracker or forums system (either way ideally gratis >>> and hosted) to complement Uservoice. Suggestions? >> It would make sense to find a tracker that both users and devs can use. >> Saves the overhead of moving things from e.g. a forums system to a bug >> tracker. > > Is it possible? Is Trac something that end users can use? I don't think Trac is the perfect solution (not as it is right now, at least), but it seems much better than our current solution (Mantis + Wikka Wakka). Pidgin (see http://developer.pidgin.im/) has an interesting implementation directly into their website. The bar at the top contains easy access to some of the most used features: Wiki (starts here), Timeline (aka "what's new?"), Roadmap and Search. It is possible to create other things like "New ticket" and "Browse source" it seems, if you look at the official trac site (http://trac.edgewall.org/). I don't have any personal experience with Trac though. Perhaps someone else around here has, and can give us some recommendations? - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Update on getting rid of emu
bo-le wrote: > Am Samstag, 31. Oktober 2009 17:28:38 schrieb Zero3: >> Matthew Toseland wrote: >>> On Friday 30 October 2009 17:10:02 Zero3 wrote: >>>> Matthew Toseland wrote: >>>>> If this line of reasoning is correct, we need to choose an >>>>> end-user-oriented issue tracker or forums system (either way ideally >>>>> gratis and hosted) to complement Uservoice. Suggestions? >>>> It would make sense to find a tracker that both users and devs can use. >>>> Saves the overhead of moving things from e.g. a forums system to a bug >>>> tracker. >>> Is it possible? Is Trac something that end users can use? >> I don't think Trac is the perfect solution (not as it is right now, at >> least), but it seems much better than our current solution (Mantis + >> Wikka Wakka). >> >> Pidgin (see http://developer.pidgin.im/) has an interesting >> implementation directly into their website. The bar at the top contains >> easy access to some of the most used features: Wiki (starts here), >> Timeline (aka "what's new?"), Roadmap and Search. It is possible to >> create other things like "New ticket" and "Browse source" it seems, if >> you look at the official trac site (http://trac.edgewall.org/). >> >> I don't have any personal experience with Trac though. Perhaps someone >> else around here has, and can give us some recommendations? >> > this may fit our needs better then trac: http://basieproject.org/ A third option is Google Project Hosting (we are already using some Google-thingy for downloads I think?). Example here: http://code.google.com/p/support/issues/list The interface seems quite simple, and has both bug tracker and wiki. I'm quite fond of the "starring" of bugs. It's basically the possibility for users to mark the bugs they are specially interested in, which also gives the developers the possibility to focus on the most popular bugs. This might be able to supercede uservoice too (which is quite prone to spam as no user verification of votes are done). - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Freenet Summer of Code wrap-up
Matthew Toseland wrote: > Apologies for being the absolute last wrap-up! This year went really well, we > had 5 students, and they all (more or less) deserved their passes. We had a > much stronger selection process than in the first 2 years, requiring some > demonstration in the form of code: a new feature or a bugfix. So even though > most of the students were completely new to us, mostly they did pretty well. > One of our students had a work conflict, but this was resolved satisfactorily. > > For the Googlers reading this, Freenet is an anonymous peer to peer system > with support for forums, browsing the internal web, filesharing etc, with a > focus on security and the option of running in "darknet" or friend-to-friend > mode. It is intended (at least by me) for people in hostile environments > (China, Iran etc) to express themselves freely, but it is currently mostly > used in the West by geeks, filesharers, etc. > > infinity0 and mikeb worked together on a new searching plugin, which we have > now deployed. infinity0's work was primarily on a new index format (which > works, but the spider needs more work), and on distributed indexing (which > doesn't yet), and mikeb mainly worked on improving the user interface and > adding essential features (simple non-infringing page ranking algorithm, > booleans, phrase matches etc). kurmi worked on new filtering code for various > formats, particularly a vastly improved CSS filter, which needed considerable > work to sort out all the parsing perversities but is now merged (Freenet has > to be very careful not to send anything to the browser that might give the > user's IP away via a web-bug). ljb worked on more friend-to-friend > functionality, his work is included in current builds. sashee worked on > making the web interface more dynamic, including solving a long-running > problem with image loading blocking the browser (freenet has quite high > latency!), using Google Web Toolkit; this branch has not yet been merged, but hopefully will be inside the next 6 months or so, it needs some tuning and debugging for slow browsers. > > Some of our students achieved less than expected IMHO but in more cases there > was a lot more work involved in the project than I expected, and the students > did really well. I have talked to most of them in the last month, well after > the programme was finished, and hopefully some of them will continue to > contribute at least occasionally. Best year yet, many thanks to Google! /me *claps* - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] [FYI] Sun Java SE 5.0 EOL today.
Ian Clarke wrote: > That being said, I think the key question is still: what do we gain by > dropping 1.5 compatibility? IMHO, for what it's worth: I'm not qualified to tell exactly what kind of new stuff 1.6 introduced, but I don't think we should fall behind upstream. If 1.5 is now end-of-life, we should already have moved to 1.6, which has been out since late 2006. I don't think supporting a minority of users with hopelessly outdated systems justifies holding back development. - Zero ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Detecting the language in the installer
Matthew Toseland wrote: > Does autodetecting the language from the OS work in the wininstaller now? If > not, what do we need to do to make it work? What kind of language codes will > Windows return us? Autodetecting has always worked in the wininstaller, but the setting passed on to fproxy has been more or less guesswork. At the moment the localized name of the language will be passed on via freenet.ini. This is highly unsuitable and will only work in some lucky cases. Windows gives us a Windows locale code consisting of 4 hexadecimal characters (see http://www.autohotkey.com/docs/misc/Languages.htm). To make it work, fproxy should either accept these Windows locale codes or someone should build a mapping file that can map these codes to something else. - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Detecting the language in the installer
Matthew Toseland wrote: > I will make fproxy accept the windows locales. Should they be preceded with a > prefix like WINDOWS0409? Cool. Whatever you want :). The 4 chars alone are fine with me too. - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Some feedback from a hostile environment
Matthew Toseland wrote: > Our friend has also localised the wininstaller (this is subject to technical > issues Zero3 hopefully will be able to resolve), and jSite (I will deal with > this soon). Yah. If anyone has a Windows setup with a cyrillic/chinese/japanese/... locale, I'd be grateful if they would do a little testing for me. - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Some feedback from a hostile environment
Matthew Toseland wrote: > https://bugs.freenetproject.org/view.php?id=3697 > Zero3, any chance of writing the total system memory to a file in the > wininstaller in the near future? No problem. 1) I assume we are talking about physical memory here? 2) Do we look at total memory available or memory currently free? 3) Exactly what should it write, and where? 4) Shall I implement this in the installer (only set once) or the launcher? (set every time user launches Freenet via our shortcuts) 5) Would it be a good idea to deny installation if the user doesn't meet the minimum requirements? - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
[freenet-dev] Website downloads page feedback
Feedback for the website downloads page: "We strongly recommend you use the links on the start menu to shut down Freenet when you play computer games, and start it back up afterwards. We will implement a system tray icon for this in future." This is implemented now. "Freenet works best with Windows XP Professional. Freenet will run on Vista and Windows 7 but the uninstaller is not perfect on those systems." The wininstaller works smoothly on both XP and Vista now the beta has been rolled out. Windows 7 is *not* supported at this time. I've seen several Win7 bug reports, so until I have time to look at Win7 support we should discourage using Freenet on Win7. "You can access Freenet later on via the Browse Freenet icon on the desktop and your start menu. If the browser window isn't opened, " On Windows, the main method of accessing Freenet is now via the tray menu. Alternatively the start menu. We don't add a desktop shortcut by default any more. " for example because you used the headless installer" Most people won't know what a "headless installer" is. "Recommended: 1GHz or more processor with 512MB RAM or more (especially if using Windows XP)" As opposed to? Win2k? Vista? Win7? "Windows users can upgrade to the latest-stable Freenet release by clicking on "update.cmd" in the Freenet directory." On Windows, this is now implemented in the tray manager. - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] Website downloads page feedback
Matthew Toseland wrote: > On Tuesday 10 November 2009 23:12:03 Zero3 wrote: >> Feedback for the website downloads page: >> >> "We strongly recommend you use the links on the start menu to shut down >> Freenet when you play computer games, and start it back up afterwards. >> We will implement a system tray icon for this in future." >> >> This is implemented now. >> >> "Freenet works best with Windows XP Professional. Freenet will run on >> Vista and Windows 7 but the uninstaller is not perfect on those systems." >> >> The wininstaller works smoothly on both XP and Vista now the beta has >> been rolled out. >> >> Windows 7 is *not* supported at this time. I've seen several Win7 bug >> reports, so until I have time to look at Win7 support we should >> discourage using Freenet on Win7. >> >> "You can access Freenet later on via the Browse Freenet icon on the >> desktop and your start menu. If the browser window isn't opened, " >> >> On Windows, the main method of accessing Freenet is now via the tray >> menu. Alternatively the start menu. We don't add a desktop shortcut by >> default any more. >> >> " for example because you used the headless installer" >> >> Most people won't know what a "headless installer" is. >> >> "Recommended: 1GHz or more processor with 512MB RAM or more (especially >> if using Windows XP)" >> >> As opposed to? Win2k? Vista? Win7? >> >> "Windows users can upgrade to the latest-stable Freenet release by >> clicking on "update.cmd" in the Freenet directory." >> >> On Windows, this is now implemented in the tray manager. > > So we probably need to split it up by OS? Or maybe not ... > > We do need to mention that on 64-bit Windows we use a 32-bit JVM which may > not be auto-updated, since this trips people up from time to time? Until Linux and Mac has tray support, I guess we should split it up. Yea, if that is the case (i can't figure out what is up and down with this 32-bit vs. 64-bit stuff), we should mention that people with 64-bit Windows without Java will be offered to install a 32-bit Java. - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] selecting browser for Launch Freenet from systray
Ed Tomlinson wrote: > On Saturday 14 November 2009 16:02:11 Matthew Toseland wrote: >> On Saturday 14 November 2009 17:19:15 Ed Tomlinson wrote: >>> Hi, >>> >>> For a test I installed freenet on a windows 7 box. All went well. The new >>> installation process is pretty simple. One small bone to pick. The first >>> use wizard tells you NOT to use the same browser for freenet as you use >>> normally - fine. However there does not seem to be a way to configure >>> 'Launch Freenet' from the systray to use your prefered browser. I want to >>> set it to use chrome... >> It *does* use Chrome if available, in incognito mode. Ask Zero3 why it isn't >> working. >>> This should be easy but it does not seem to be... > > Think the problem is that I installed chrome after freenet and it _is_ what I > want freenet to use. I see no option to tell freenet to switch to it. Thanks for the feedback :) Here is the deal: At the moment, there is a great difference in security between the major browsers. Especially their incognito mode support. For that reason, the launcher will dictate the choice of browser for the user by trying to find the most secure one available on every launch. If you install Chrome, the launcher should pick it up at next launch. More specifically, it will look for the registry string "InstallLocation" under "HKEY_CURRENT_USER, Software\Microsoft\Windows\CurrentVersion\Uninstall\Google Chrome". Can you check if you have that key? The fproxy message is a bit old. It was made at a time where the plan was to install a secondary browser/browser profile for the user, and use that one solely for Freenet. As browsers in the meantime have started to incorporate incognito modes, we are moving towards using those instead (and IMHO, it is a much better solution. Last time we tried the profile stuff - with FireFox - we failed miserably). I'm surprised you were able to install on Win7. I haven't tested the installer on Win7 myself yet, and I've seen reports of installations failures on Win7. It just worked out of the box? - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] selecting browser for Launch Freenet from systray
Ed Tomlinson wrote: > On Saturday 14 November 2009 21:19:43 Zero3 wrote: >> Ed Tomlinson wrote: >> If you install Chrome, the launcher should pick it up at next launch. More specifically, it will look for the registry string "InstallLocation" under "HKEY_CURRENT_USER, Software\Microsoft\Windows\CurrentVersion\Uninstall\Google Chrome". Can you check if you have that key? >> Does not exist. I installed the 64 bit version. The uninstall entry is at: > HKEY_LOCAL_MACHINE\SOFTWARE\Wow6432\Microsoft\Windows\CurrentVersion\Unistall\Google Chrome Argh. Why on earth did they change that? Is that true for all versions of 64-bit Windows? Win7? Vista? (XP?). >> I'm surprised you were able to install on Win7. I haven't tested the installer on Win7 myself yet, and I've seen reports of installations failures on Win7. It just worked out of the box? > > Aside from the chrome issue it worked as expected. Cool :o. - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] selecting browser for Launch Freenet from systray
Matthew Toseland wrote: > On Sunday 15 November 2009 01:14:54 Ed Tomlinson wrote: >> On Saturday 14 November 2009 16:02:11 Matthew Toseland wrote: >>> On Saturday 14 November 2009 17:19:15 Ed Tomlinson wrote: >>>> Hi, >>>> >>>> For a test I installed freenet on a windows 7 box. All went well. The >>>> new installation process is pretty simple. One small bone to pick. The >>>> first use wizard tells you NOT to use the same browser for freenet as you >>>> use normally - fine. However there does not seem to be a way to configure >>>> 'Launch Freenet' from the systray to use your prefered browser. I want to >>>> set it to use chrome... >>> It *does* use Chrome if available, in incognito mode. Ask Zero3 why it >>> isn't working. >>>> This should be easy but it does not seem to be... >> Think the problem is that I installed chrome after freenet and it _is_ what >> I want freenet to use. I see no option to tell freenet to switch to it. > > That is odd, I thought it detected it in the launcher at launch time? Have > you filed a bug about this? It does. If you read his previous reply, you will see that Microsoft changed the registry path for uninstall entries in 64-bit Win7 (and possibly 64-bit Vista/XP too?). Which means the launcher won't find Chrome on these systems... I need to figure out which versions that use the changed path, and change the launcher for these versions. - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
Re: [freenet-dev] selecting browser for Launch Freenet from systray
Matthew Toseland wrote: > On Tuesday 17 November 2009 07:20:03 Zero3 wrote: >> Matthew Toseland wrote: >>> On Sunday 15 November 2009 01:14:54 Ed Tomlinson wrote: >>>> On Saturday 14 November 2009 16:02:11 Matthew Toseland wrote: >>>>> On Saturday 14 November 2009 17:19:15 Ed Tomlinson wrote: >>>>>> Hi, >>>>>> >>>>>> For a test I installed freenet on a windows 7 box. All went well. The >>>>>> new installation process is pretty simple. One small bone to pick. The >>>>>> first use wizard tells you NOT to use the same browser for freenet as >>>>>> you use normally - fine. However there does not seem to be a way to >>>>>> configure 'Launch Freenet' from the systray to use your prefered >>>>>> browser. I want to set it to use chrome... >>>>> It *does* use Chrome if available, in incognito mode. Ask Zero3 why it >>>>> isn't working. >>>>>> This should be easy but it does not seem to be... >>>> Think the problem is that I installed chrome after freenet and it _is_ >>>> what I want freenet to use. I see no option to tell freenet to switch to >>>> it. >>> That is odd, I thought it detected it in the launcher at launch time? Have >>> you filed a bug about this? >> It does. If you read his previous reply, you will see that Microsoft >> changed the registry path for uninstall entries in 64-bit Win7 (and >> possibly 64-bit Vista/XP too?). Which means the launcher won't find >> Chrome on these systems... >> >> I need to figure out which versions that use the changed path, and >> change the launcher for these versions. > > Looks like that's simply due to installing a 32-bit Chrome on a 64-bit > Windows... Hmm. Maybe. But that would mean that our own uninstallation entry would be "relocated" on those systems as well. Which means uninstallation might not work as predicted. In any way, I need to sort out exactly what happens - and when it happens. - Zero3 ___ Devl mailing list Devl@freenetproject.org http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
[freenet-dev] financial status
Matthew Toseland skrev: > According to Ian, we now have $1,700 in the paypal account! Looks like > putting > the balance on the homepage has helped! > Congrats :) > Input would be appreciated on the last change I made to the wording. It now > says: > "The project's current balance is $6519.72. > We have enough money to pay for the project's full time developer and the > server for around another 78 days. If you would like to help support the > Freenet Project, click here to make a donation." > > "around another 78 days" has a tooltip which explains: > approximate figure not including work already paid for (1 month or less) > > Is "around another 78 days" readable? Should we explain it in a bracket? > Should we get rid of the tooltip? Etc. > Personally, I'd word it like (just a suggestion) "With this amount, we can fund the continued development of the Freenet project for approximately XX days.". Details such as "not including work already paid for" are not really that important, so scrap the tooltip IMO. - Zero3
[freenet-dev] financial status
Matthew Toseland skrev: >>> "The project's current balance is $6519.72. >>> We have enough money to pay for the project's full time developer and the >>> server for around another 78 days. If you would like to help support the >>> Freenet Project, click here to make a donation." >>> >>> "around another 78 days" has a tooltip which explains: >>> approximate figure not including work already paid for (1 month or less) >>> >>> Is "around another 78 days" readable? Should we explain it in a bracket? >>> Should we get rid of the tooltip? Etc. >>> >>> >> Personally, I'd word it like (just a suggestion) "With this amount, we >> can fund the continued development of the Freenet project for >> approximately XX days.". Details such as "not including work already >> paid for" are not really that important, so scrap the tooltip IMO. >> > > Shouldn't we be rigorously honest, within reason? > I agree with that principle. To me, that disclaimer doesn't seem important though. Whether or not it is, is a matter of personal opinion I guess. Not my call, just throwing feedback :) - Zero3
[freenet-dev] State of the Freenet, and tentative roadmap
Thanks for the update - nice to know what's going on! - Zero3
[freenet-dev] Installation -
Ian Clarke skrev: > http://www.littleshoot.org/ > > While Nextgens has done great work improving the installation process, > clearly there is plenty of room for further improvement. I see no > reason why Freenet's installation process couldn't be as elegant as > littleshoot's. > I haven't tried out the LittleShoot installer (yet), but I also do agree that the installer contains unnecessary steps (I did point out a few of them in my review some time ago). IMHO, Freenet should be packaged for Linux distros anyway, compeletely eliminating the need for a Linux installer (besides whatever scripting the package installation requires) and terminal installation procedures. I am aware of certain people thinking that Freenet develops too quickly/doesn't fit into package systems at all, but the advantages still seem to be much greater than the disadvantages as I see it. I do see the reasoning behind using IzPack (isn't that the name?) because of cross-platform support though, but assuming Linux is the future, and Linux apps ought to be packaged anyway, we only have Windows and Mac left, leaving less reason to be bound to the, perhaps less intuitive, IzPack installation? - Zero3
[freenet-dev] Installation -
Ian Clarke skrev: > On Tue, Nov 25, 2008 at 3:04 AM, Zero3 wrote: > >> but assuming Linux is the future, and Linux apps ought to be packaged >> anyway, we only have Windows >> and Mac left, leaving less reason >> > > An unwarranted assumption. > True. But don't you agree that Linux is gaining market share at the moment, that these new users prefer the easy GUI-based distros like Ubuntu, and the de-facto standard of installing software on these distros are via packaging systems? Getting Freenet packaged (and prepared to be such properly) and available in the Debian/Ubuntu repository seems like a great step towards a more user-friendly installation on Freenet on Linux. A quick roundup on advantages/features for Freenet (just to sum up): - One-click installation (any real settings should be handled by the first-time wizard) - Integration directly into the OS via "Add/Remove programs" (in Ubuntu, for example) and the package manager (which also means free publicity and more users) - Automatic, fail-safe downloading and updating with checksum and signature checking (no need for the manual update scripts and maintaining them) - Less maintenance for the installation maintainer - Depency handling (e.g. require a specific version of the Java runtime and other required libraries) - ... probably more if you think about it. > As open source fans we all want to see Linux do well on the desktop, > but we can't allow our hopes for Linux to lead to sub-optimal decision > making when it comes to maximizing Freenet's adoption. Windows *is* > the most important OS for Freenet adoption. Whether Macs or Linux > come next is up for debate, but at the very least, a simple and > elegant installation is important on all three platforms. > Really what we need are dedicated maintainers for the installers on > Windows, Mac, and perhaps a few of the major Linux distros. An > installer that works on all three platforms has many advantages, but > will never be as smooth or intuitive as platform-specific installers > because people have differing expectations of each platform. For > example, Windows users tend to expect a Wizard-style installer. Mac > users expect a DMG containing an executable App that they can drag to > their Applications folder. Linux users expect to be able to use > apt-get, yum, or something else depending on their specific distro. > I completely agree about the importance of Windows users and the appeal and simplicity of the installation procedure in order to meet the user demands. The reasons for sticking with the current installation procedure are, if I remember correctly, easiness (if that's even a word) to maintain and cross-platform support. While this might work "okay" at the moment, I personally think splitting the installation process up to be platform-specific is the way to go - especially because of the huge differences in installation procedures between the platforms, as you mentioned. > Next, we must identify anything that can be improved in Freenet that > would make writing these installers easier. > Random example on top of my head is the downloading of the plugins *during* the actual installation process, from the Freenet website. Surely they ought to be packed into the installer next to the other files? (The question on whether to *use* the various plugins ought to be asked during the first-time wizard IMHO. Atm. it seems like all installed plugins are automatically loaded?) To make proper packaging possible on Linux, and to properly support multi-user environments on Windows, Freenet also needs to separate the identity, machine settings and eventually the cache, from the program files. Quick example of where different things probably ought to be located in Windows and Linux (sorry, don't know anything about Macs): - Windows: Program files: "%ProgramFiles%\Freenet" Machine-specific settings: Same location as program files, or in "%ALLUSERSPROFILE%\Application Data\Freenet", or if run as a service, in the service's appdata folder. User-specific data: "%AppData%\Freenet" - Linux: Program files: "/usr/lib/freenet" (and executable in "/usr/bin"?) Machine-specific settings: "/etc/freenet" User-specific data: "~/.freenet" At the moment, everything is throw into "%ProgramFiles%\Freenet" on Windows and "~/.Freenet" on Linux. Another thing to solve is the disagreements on how Freenet should operate on Windows. Atm. Freenet creates its own user account and installs itself as a service, as opposed to running as a normal background application as the logged in user. - Zero3
[freenet-dev] Installation -
Florent Daigniere skrev: > Zero3 wrote: > >> >> IMHO, Freenet should be packaged for Linux distros anyway, compeletely >> eliminating the need for a Linux installer (besides whatever scripting >> the package installation requires) and terminal installation procedures. >> > > Send a patch. > > Is that meant as a "i-don't-care-smart-ass-comment" or a friendly suggestion? >> I am aware of certain people thinking that Freenet develops too >> quickly/doesn't fit into package systems at all, but the advantages >> still seem to be much greater than the disadvantages as I see it. >> >> > > We all agree that packaging makes sense... but they are two problems in > freenet's case: > 1- the source code evolves too fast, meaning that the packages will be > unsuitable to be included in the main distro's repositories... That > means that the user *will* have to do something to his packaging system > to install freenet. On debian that would mean adding a new repository: > arguably that's not simple and it requires r00t priviledges, which isn't > the case of the current installer. > 2- Lack of manpower: It's way faster/easier to maintain ONE platform > agnostic installer than N packages for N distributions. > 1) Surely it is possible to package connectable version of Freenet which will automatically update itself over Freenet (or perhaps a fallback to update over http)? (If package managers won't allow you to override packaged files, update to an "overlay" directory where any files will transparently override the packaged files) 2) Once you have created the packages and take advantage of the features of package management, it seems easier to me. But that's a hard thing to argue both for or against. > >> I do see the reasoning behind using IzPack (isn't that the name?) >> because of cross-platform support though, but assuming Linux is the >> future, and Linux apps ought to be packaged anyway, we only have Windows >> and Mac left, leaving less reason to be bound to the, perhaps less >> intuitive, IzPack installation? >> >> - Zero3 >> > > Huh. See my other emails... > Which? (Sorry if I missed any relevant) - Zero3
[freenet-dev] Installation -
Florent Daigniere skrev: > Zero3 wrote: > >> Ian Clarke skrev: >> >>> On Tue, Nov 25, 2008 at 3:04 AM, Zero3 wrote: >>> >>> >>>> but assuming Linux is the future, and Linux apps ought to be packaged >>>> anyway, we only have Windows >>>> and Mac left, leaving less reason >>>> >>>> >>> An unwarranted assumption. >>> >>> >> True. But don't you agree that Linux is gaining market share at the >> moment, that these new users prefer the easy GUI-based distros like >> Ubuntu, and the de-facto standard of installing software on these >> distros are via packaging systems? Getting Freenet packaged (and >> prepared to be such properly) and available in the Debian/Ubuntu >> repository seems like a great step towards a more user-friendly >> installation on Freenet on Linux. A quick roundup on advantages/features >> for Freenet (just to sum up): >> >> - One-click installation (any real settings should be handled by the >> first-time wizard) >> > > Then it would require the node to have web-access and to make > web-requests after it has been set up. The current node doesn't do that > unless told to. > Web access for what? > >> - Integration directly into the OS via "Add/Remove programs" (in Ubuntu, >> for example) and the package manager (which also means free publicity >> and more users) >> > > Can't be done if we aren't in the main repositories. > ... isn't the point to get into the main repositories at some point? I'm not sure if you can get programs in that list form 3rd party repositories (for any distros without official packages). It might be possible. > >> - Automatic, fail-safe downloading and updating with checksum and >> signature checking (no need for the manual update scripts and >> maintaining them) >> > > I'm not sure I understand what you mean here. > I was merely trying to point out some of the technical advantages of packaging systems - here referring to the actual package download and updating part. The update scripts deal with the downloading and checksumming/verification "manually" atm. With a package, there shouldn't be a need for any update scripts or worrying about genuine downloads in the first place (which means less clutter, less manual work and fewer places things can go wrong). > >> - Less maintenance for the installation maintainer >> > > Neither here... the idea is to outsource the maintenance of the > installer, isn't it? > Meant as: "Less work for *whoever* does the installer/package stuff". > >> - Depency handling (e.g. require a specific version of the Java runtime >> and other required libraries) >> - ... probably more if you think about it. >> >>> As open source fans we all want to see Linux do well on the desktop, >>> but we can't allow our hopes for Linux to lead to sub-optimal decision >>> making when it comes to maximizing Freenet's adoption. Windows *is* >>> the most important OS for Freenet adoption. Whether Macs or Linux >>> come next is up for debate, but at the very least, a simple and >>> elegant installation is important on all three platforms. >>> Really what we need are dedicated maintainers for the installers on >>> Windows, Mac, and perhaps a few of the major Linux distros. An >>> installer that works on all three platforms has many advantages, but >>> will never be as smooth or intuitive as platform-specific installers >>> because people have differing expectations of each platform. For >>> example, Windows users tend to expect a Wizard-style installer. Mac >>> users expect a DMG containing an executable App that they can drag to >>> their Applications folder. Linux users expect to be able to use >>> apt-get, yum, or something else depending on their specific distro. >>> >>> >> I completely agree about the importance of Windows users and the appeal >> and simplicity of the installation procedure in order to meet the user >> demands. The reasons for sticking with the current installation >> procedure are, if I remember correctly, easiness (if that's even a word) >> to maintain and cross-platform support. While this might work "okay" at >> the moment, I personally think splitting the installation process up to >> be platform-specific is the way to go - especially because of the huge
[freenet-dev] Laptops still suck was Re: Installation -
Matthew Toseland skrev: >> or not to use the >> network is an open question. Freenet as currently implemented doesn't >> play nice with laptops... maybe we should be more clear about that on >> the website. >> > > With big flashing warning signs? I wonder if there's any way to detect that > the user is running a laptop pre-install... or post-install for that > matter... :) Anyway even on a desktop we will still have poor uptime, so I > guess there's little point in nagging the user about it. > Check for a battery? ;) > To clarify for anyone who hasn't got the picture yet: > Low uptime is very bad for Freenet. > Low uptime darknet is nearly impossible for Freenet. > Poor connectivity is bad for Freenet. > Uncontrolled NATs and mobile nodes are bad for Freenet. > > To sum it up: > LAPTOPS ARE BAD FOR FREENET! > > 0.10, as currently planned, will help a bit, but even so, uptime is always > going to be a serious problem... Should we show a flashing warning sign if > our uptime is below some percentage? > > Shouldn't we just accept that fact that people are moving towards mobility, and that most likely won't change anytime soon? :). IPv6 will probably solve most of the NAT and IP issues with time. I doubt nagging the user about uptime will help anything. Do we really want to support people keeping their laptops online just to seed the Freenet network, compared to saving power (you know, lack of oil in the world, money for electricity, global heating, etc.) by turning it off? Can we morally justify that? Is Freenet that important? (I'm not saying it is or isn't, but think about it for a minute) - Zero3
[freenet-dev] Moving more stuff into the wizard; geeks vs users was Re: Installation -
Matthew Toseland skrev: >> An >> installer that works on all three platforms has many advantages, but >> will never be as smooth or intuitive as platform-specific installers >> because people have differing expectations of each platform. For >> example, Windows users tend to expect a Wizard-style installer. Mac >> users expect a DMG containing an executable App that they can drag to >> their Applications folder. Linux users expect to be able to use >> apt-get, yum, or something else depending on their specific distro. >> > > Unless their specific distro happens to be unsupported. Which is common, > because the distro market is still extremely fragmented. Hence we need a good > GUI installer even for linux. No? > deb and rpm probably covers most of the GUI distros. The "Alien" program can convert packages to various other formats if needed. >> Next, we must identify anything that can be improved in Freenet that >> would make writing these installers easier. >> > > IMHO moving the "wizard" part into the node itself was an important step in > the right direction. We could move the rest into the node by always > downloading the plugins and seednodes file in the installer, and asking the > user about the plugins during the post-install wizard. Ideally we'd also ask > the user about auto-start in the post-install wizard (defaulting on but > executing a script to turn it off if the user asks us to). > I agree. It doesn't seem like that big of a task to move the rest of the stuff into the wizard (now you already have the framework). I'd argue for a default autorun status of "off" though, and instead make "on" the default on the relevant wizard page, along with a short description of advantages/consequences of autorun'ing Freenet. That way we mess as little as possible with the user's system without asking for permission first - and considering how many resources Freenet uses, I think that's more than fair. Almost every user will simply accept the default autorun settings and continue, and those who doesn't want Freenet to run automatically will get their fair choice to say "no" without feeling stepped onto. - Zero3
[freenet-dev] Packaging: Linux, Windows issues was Re: Installation -
> can simply create a user-specific startup item. This would certainly simplify > installation. > Maybe we should ask the user? But that means more steps in the installation > process!! > If we (as previously discussed) move the autorun question to the wizard and separate the identity from the program files, we can simply add Freenet to each user's startup group as we are given permission to from the users (assuming every user will get to complete their own copy of the first-time wizard). Maximum userfriendliness... >> Machine-specific settings: "/etc/freenet" >> User-specific data: "~/.freenet" >> > > IMHO there is no user-specific data for Freenet. It's a daemon, like Apache. > It may implement its own user mechanism, > The identity? Friends (darknet)? fproxy theme and similar settings? All seems user-specific to me? - Zero3
[freenet-dev] Laptops still suck was Re: Installation -
Florent Daigni?re skrev: > * Zero3 [2008-11-25 23:55:20]: > > >> Shouldn't we just accept that fact that people are moving towards >> mobility, and that most likely won't change anytime soon? :). >> >> > > I would really appreciate if you could share your insight on the topic: > How exactly would you design a fully decentralized protocol where most > of the nodes are hardly ever connected and mostly unreachable? > I'm not saying I have the magical answers, I'm just pointing out that denying that the world is changing the way it is won't create any solutions. With my basic knowledge of this stuff, I'm thinking: "less uptime means people need more peers". I got the impression that the peering of nodes is far from as trivial and automated as it could be. Perhaps a solution could be integration of external WoT's into Freenet. I'm not familiar with how WoT works for social platforms like Facebook, but I imagine you could create some kind of plugin to transfer (trusted) friends from social platforms into Freenet nodes' darknet peer lists (yes, I am aware of the privacy and security concerns and the challenges involved - I'm just brainstorming a possible solution.). - Zero3
[freenet-dev] Laptops still suck was Re: Installation -
Florent Daigni?re skrev: >> I'd argue for a default autorun status of "off" though, and instead make >> "on" the default on the relevant wizard page, along with a short >> description of advantages/consequences of autorun'ing Freenet. >> > > Here you are just being silly... but let's address that on the other > thread: > http://archives.freenetproject.org/message/20081126.001810.5c2fe629.en.html > Silly? I disagree. Giving the user the choice to decide over how his own system runs doesn't seem silly to me at all. Not automatically autorunning without the user's accept to do so seems like the proper way (even though much software goes against that practice). Informing about the performance consequences really is in place too, even if it causes some user to disable the autorun. Or not? - Zero3
[freenet-dev] Moving more stuff into the wizard; geeks vs users was Re: Installation -
Florent Daigni?re skrev: > * Zero3 [2008-11-26 00:08:17]: > > >> Matthew Toseland skrev: >> >>>> An >>>> installer that works on all three platforms has many advantages, but >>>> will never be as smooth or intuitive as platform-specific installers >>>> because people have differing expectations of each platform. For >>>> example, Windows users tend to expect a Wizard-style installer. Mac >>>> users expect a DMG containing an executable App that they can drag to >>>> their Applications folder. Linux users expect to be able to use >>>> apt-get, yum, or something else depending on their specific distro. >>>> >>>> >>> Unless their specific distro happens to be unsupported. Which is common, >>> because the distro market is still extremely fragmented. Hence we need a >>> good >>> GUI installer even for linux. No? >>> >>> >> deb and rpm probably covers most of the GUI distros. The "Alien" program can >> convert packages to various other formats if needed. >> >> > > That's not proper packaging. > > If the converted packages are just as good as manually ported? (I don't know if they are, all I know is that alien is available and that's what it's supposed to do) >>>> Next, we must identify anything that can be improved in Freenet that >>>> would make writing these installers easier. >>>> >>>> >>> IMHO moving the "wizard" part into the node itself was an important step in >>> the right direction. We could move the rest into the node by always >>> downloading the plugins and seednodes file in the installer, and asking the >>> user about the plugins during the post-install wizard. Ideally we'd also >>> ask >>> the user about auto-start in the post-install wizard (defaulting on but >>> executing a script to turn it off if the user asks us to). >>> >>> >> I agree. It doesn't seem like that big of a task to move the rest of the >> stuff into the wizard (now you already have the framework). >> >> > > Putting stuffs in the wizard goes against the packaging logic. On debian > you would want to use debconf to ask the user on how to configure his > node... > > Both ways should probably be supported. Debian(-like) packages could ask for answers via debconf, and the wizard could take over if settings was not set via debconf. Other distros might have similar methods? Seems like the proper way to support all Linux distros? - Zero3
[freenet-dev] Installation -
Florent Daigni?re skrev: > * Zero3 [2008-11-25 23:49:24]: > > >>> Then it would require the node to have web-access and to make >>> web-requests after it has been set up. The current node doesn't do that >>> unless told to. >>> >>> >> Web access for what? >> >> > > Downloading plugins. > Assuming we are not packaging them with Freenet... Even if we don't, does it matter that much if it is the installer or the node that makes the request? Matter more than having a true one-click installation? >>>> - Integration directly into the OS via "Add/Remove programs" (in Ubuntu, >>>> for example) and the package manager (which also means free publicity >>>> and more users) >>>> >>>> >>> Can't be done if we aren't in the main repositories. >>> >>> >> ... isn't the point to get into the main repositories at some point? I'm >> not sure if you can get programs in that list form 3rd party >> repositories (for any distros without official packages). It might be >> possible. >> >> > > We can't because of the frequency of required updates... and because > our code depends on a non-free jvm. > > See suggestion about "overlaying" updates. I didn't know that the jvm wasn't free? Will there be one available within reasonable time perhaps, or will we have to depend on the non-free one later on? >>> >>> >>>> - Automatic, fail-safe downloading and updating with checksum and >>>> signature checking (no need for the manual update scripts and >>>> maintaining them) >>>> >>>> >>> I'm not sure I understand what you mean here. >>> >>> >> I was merely trying to point out some of the technical advantages of >> packaging systems - here referring to the actual package download and >> updating part. The update scripts deal with the downloading and >> checksumming/verification "manually" atm. With a package, there >> shouldn't be a need for any update scripts or worrying about genuine >> downloads in the first place (which means less clutter, less manual work >> and fewer places things can go wrong). >> > > Ok, whatever: you're preaching a converted here. > > Okay :) >>>> - Less maintenance for the installation maintainer >>>> >>>> >>> Neither here... the idea is to outsource the maintenance of the >>> installer, isn't it? >>> >>> >> Meant as: "Less work for *whoever* does the installer/package stuff". >> >> > > Right now we have no one but me and Tommy addressing the installation-related > problems. I'd love that to change... hence I would be welcoming your > patches. > > [snip.] > > My time is quite limited (yea right, like anyone has enough...), but I'm willing to take a look at a simpler Windows installer. Making Linux packages would take me much longer as I haven't really messed around with that very much in the past. >>> The idea is to minimize the amount of data to download in order to both >>> spare bandwidth and reduce the overall installation time. >>> >>> >> Not worth the trouble/annoyances/extra download time/... IMHO. >> > > That's your view, not mine. Come back with figures and real arguments if > you plan to be convincing. Last time I checked I am the one who wrote > that part of the code... So I am the one who decides how it's done. > > That seems like an awfully closed-minded attitude for a collaborative open-source project like Freenet. Being hosted at SourceForge, I can't see bandwidth being a problem? But since you want some figures: I just did a test install. Downloading and setting up the plugins took the installer ~10 seconds on a 2 year old mainstream laptop with Windows XP. The plugins take up 383 KB. I don't know how many people that uncheck any or all of the plugins before installing, but I doubt it's a large part. Even if *everybody* unchecked all plugins in the installer and we assume nobody will ever install them later on, the overhead would be less than 4% of the ~10 MB that was downloaded during the install. In reality, that number will be *much* smaller as many people *will* install the plugins. If SourceForge can't keep up with that little extra bandwidth, I'll be glad to donate. >> If it >> real
[freenet-dev] Laptops still suck was Re: Installation -
Matthew Toseland skrev: >> IPv6 will >> probably solve most of the NAT and IP issues with time. >> > > No, first IPv6 will probably not be widely used for a lng time, secondly, > people will probably still use evil firewalls if not NATs, especially on > shared/public connections, thirdly fast mobile connections are billed per gig > and slow ones are throttled so severely as to be unusable for us, and this is > likely to remain true for some time IMHO. > "with time"... But I don't suppose there will be need for double NATs, or even NATs at all? I was of the belief that much security will be moved to the OS with IPv6 and enough addresses for every single device out there to get its own. We will probably see even more support for UPnP too? - Zero3
[freenet-dev] Installation -
Tommy[D] skrev: > Zero3 schrieb: > >> 1) Surely it is possible to package connectable version of Freenet which >> will automatically update itself over Freenet (or perhaps a fallback to >> update over http)? (If package managers won't allow you to override >> packaged files, update to an "overlay" directory where any files will >> transparently override the packaged files) >> > > Bad idea: A proberly packaged version does not ship a freenet-ext.jar, but > instead packages for the > content of that jar. If you now update freenet.jar to a version, which needs > a new feature from > freenet-ext.jar while the maintainer did not update freenet.jar nor the other > packages, freenet will > fail to start, will corrupt your database or other bad things. Hence my suggestion about "overlaying" updates on top of the packaged files until the packages are updated (obviously not the prettiest way to do things, but might be necessary until Freenet development is more stable (e.g. less mandantory updates with short warning)) :). Won't be a problem if it's our own repositories though. - Zero3
[freenet-dev] Laptops still suck was Re: Installation -
Florent Daigni?re skrev: > * Zero3 [2008-11-26 20:51:14]: > > >> Florent Daigni?re skrev: >> >>>> I'd argue for a default autorun status of "off" though, and instead >>>> make "on" the default on the relevant wizard page, along with a short >>>> description of advantages/consequences of autorun'ing Freenet. >>>> >>>> >>> Here you are just being silly... but let's address that on the other >>> thread: >>> http://archives.freenetproject.org/message/20081126.001810.5c2fe629.en.html >>> >>> >> Silly? I disagree. Giving the user the choice to decide over how his own >> system runs doesn't seem silly to me at all. Not automatically >> autorunning without the user's accept to do so seems like the proper way >> (even though much software goes against that practice). >> > > The aim was to reduce the number of questions we ask during the > installation to a minimum: on the basis that advanced users can change > the settings they need afterwards, including whether the node > auto-starts or not. > > Indeed, yet that is one of the questions you probably *ought* to ask. >> Informing about >> the performance consequences really is in place too, even if it causes >> some user to disable the autorun. Or not? >> > > "parse error" > ? - Zero3
[freenet-dev] Moving more stuff into the wizard; geeks vs users was Re: Installation -
Florent Daigni?re skrev: > * Zero3 [2008-11-26 20:51:31]: > > >> Florent Daigni?re skrev: >> >>> * Zero3 [2008-11-26 00:08:17]: >>> >>> >>> >>>> Matthew Toseland skrev: >>>> >>>> >>>>>> An installer that works on all three platforms has many >>>>>> advantages, but >>>>>> will never be as smooth or intuitive as platform-specific installers >>>>>> because people have differing expectations of each platform. For >>>>>> example, Windows users tend to expect a Wizard-style installer. Mac >>>>>> users expect a DMG containing an executable App that they can drag to >>>>>> their Applications folder. Linux users expect to be able to use >>>>>> apt-get, yum, or something else depending on their specific distro. >>>>>> >>>>>> >>>>> Unless their specific distro happens to be unsupported. Which is >>>>> common, because the distro market is still extremely fragmented. >>>>> Hence we need a good GUI installer even for linux. No? >>>>> >>>>> >>>> deb and rpm probably covers most of the GUI distros. The "Alien" program >>>> can convert packages to various other formats if needed. >>>> >>>> >>>> >>> That's not proper packaging. >>> >>> >>> >> If the converted packages are just as good as manually ported? >> > > "If" they are then there is no problem. Experience has shown they > aren't. > > Roger that. Then manually porting will be required, obviously. I'm not aware of the differences in construction between the formats though. >>>>>> Next, we must identify anything that can be improved in Freenet that >>>>>> would make writing these installers easier. >>>>>> >>>>>> >>>>> IMHO moving the "wizard" part into the node itself was an important >>>>> step in the right direction. We could move the rest into the node >>>>> by always downloading the plugins and seednodes file in the >>>>> installer, and asking the user about the plugins during the >>>>> post-install wizard. Ideally we'd also ask the user about >>>>> auto-start in the post-install wizard (defaulting on but executing >>>>> a script to turn it off if the user asks us to). >>>>> >>>>> >>>> I agree. It doesn't seem like that big of a task to move the rest of >>>> the stuff into the wizard (now you already have the framework). >>>> >>>> >>>> >>> Putting stuffs in the wizard goes against the packaging logic. On debian >>> you would want to use debconf to ask the user on how to configure his >>> node... >>> >>> >>> >> Both ways should probably be supported. >> > > I see... and how exactly is that going to reduce the maintainance cost > again? I must have missed something. > > It's more about "doing things right", really. Obviously that does require extra work in the beginning, but if you look at the big picture, I think you will save time in the end by unifying the installation procedure with the other software in the world. I won't try to convince you about that if you disagree, because that would kind of be even more a waste of time. - Zero3
[freenet-dev] Installation -
Florent Daigni?re skrev: > * Zero3 [2008-11-26 21:41:02]: > >> Florent Daigni?re skrev: >> >>> * Zero3 [2008-11-25 23:49:24]: >>> >>> >>>>> Then it would require the node to have web-access and to make >>>>> web-requests after it has been set up. The current node doesn't do >>>>> that unless told to. >>>>> >>>> Web access for what? >>>> >>>> >>> Downloading plugins. >>> >> Assuming we are not packaging them with Freenet... Even if we don't, >> does it matter that much if it is the installer or the node that >> makes the request? >> > > The node doesn't know anything about http-proxies... the installer > might at some point. > > Pack 'em in then? >> Matter more than having a true one-click installation? >> > > Yes. > Oh well. >> Will there be one available within reasonable time perhaps, or will >> we have to depend on the non-free one later on? >> >> > > Ensuring that the code works reliably on other jvms takes dev's time > we'd rather spare somewhere else. It's all a matter of priorities, like > usual. > > We all know? That didn't quite answer though: >> Will there be one available within reasonable time perhaps, and >> will we have to depend on the non-free one later on? >>>>> The idea is to minimize the amount of data to download in order to >>>>> both spare bandwidth and reduce the overall installation time. >>>>> >>>> Not worth the trouble/annoyances/extra download time/... IMHO. >>>> >>> That's your view, not mine. Come back with figures and real >>> arguments if >>> you plan to be convincing. Last time I checked I am the one who wrote >>> that part of the code... So I am the one who decides how it's done. >>> >>> >> That seems like an awfully closed-minded attitude for a >> collaborative open-source project like Freenet. >> >> Being hosted at SourceForge, I can't see bandwidth being a problem? >> >> > > We left SourceForge years ago because of their chronical unreliability. > Oh. What's http://sourceforge.net/projects/freenet/ all about then? > >> But since you want some figures: I just did a test install. >> Downloading and setting up the plugins took the installer ~10 >> seconds on a 2 year old mainstream laptop with Windows XP. The >> plugins take up 383 KB. I don't know how many people that uncheck >> any or all of the plugins before installing, but I doubt it's a >> large part. Even if *everybody* unchecked all plugins in the >> installer and we assume nobody will ever install them later on, the >> overhead would be less than 4% of the ~10 MB that was downloaded >> during the install. In reality, that number will be *much* smaller >> as many people *will* install the plugins. If SourceForge can't keep >> up with that little extra bandwidth, I'll be glad to donate. >> >> > > We did call for mirrors a while back, and we usually do before we > announce any new release. > > Right now we have 6 working ones and 13 configured. > > What are the requirements, besides standard HTTP access to the actual files? >>>> If it really matters that much, install none and let the wizard do >>>> it instead. >>>> >>> Again, that's against the packaging philosophy >>> >>> >>> >> Surely applications are allowed to ask questions on the first run? >> As does FireFox and Thunderbird, just to mention 2 large pieces of >> packaged open-source software. If they are included in the package, >> the node won't have to download them from the web. >> >> > > Neither firefox nor thunderbird do ask questions on their first run on my > debian. That's a windowsish behaviour. > > Guess it is for FF. Thunderbird asks about account information, surely? OpenOffice doesn't either? (On Windows it asks about license agreement, user account and initials). Anyway, it doesn't really change things regarding the topic... >>> >>>>> I don't get what you mean here. Are you seriously suggesting that >>>>> multi-user computers should run multiple, concurrent nodes? It's >>>>> not like run
[freenet-dev] Laptops still suck was Re: Installation -
Ian Clarke skrev: > On Wed, Nov 26, 2008 at 3:01 PM, Zero3 wrote: > >>> The aim was to reduce the number of questions we ask during the >>> installation to a minimum: on the basis that advanced users can change >>> the settings they need afterwards, including whether the node >>> auto-starts or not. >>> >>> >> Indeed, yet that is one of the questions you probably *ought* to ask. >> > > I don't know - does mysql server ask this before it installs? I think > so long as there is an option to disable, we should default to > whatever is better for the network. The user has volunteered to run > Freenet after all. Remember also that Napster probably wouldn't have > worked if they hadn't defaulted to auto-running. Obviously Freenet != > Napster, but the analogy is valid in this case. > > I get your point, but most servers with a default install don't sit in the background eating 200 MB of ram and a share of your download/upload. A default MySQL daemon is ~20 MB of ram and zero internet usage. I don't know about Napster, but IIRC it had a tray icon serving the purpose of both informing the user of the fact that it was unning, and allowing the user to easily kill it if needed. I do understand that whether the default should be on or off is a tough decision, but shouldn't we at least warn the user about it, and provide an easy way to disable it (if enabled by default)? - Zero3
[freenet-dev] Moving more stuff into the wizard; geeks vs users was Re: Installation -
Florent Daigni?re skrev: > * Zero3 [2008-11-26 22:07:11]: > >>>>>>>> Next, we must identify anything that can be improved in Freenet that >>>>>>>> would make writing these installers easier. >>>>>>>> >>>>>>>> >>>>>>> IMHO moving the "wizard" part into the node itself was an >>>>>>> important step in the right direction. We could move the rest >>>>>>> into the node by always downloading the plugins and seednodes >>>>>>> file in the installer, and asking the user about the plugins >>>>>>> during the post-install wizard. Ideally we'd also ask the user >>>>>>> about auto-start in the post-install wizard (defaulting on but >>>>>>> executing a script to turn it off if the user asks us to). >>>>>>> >>>>>>> >>>>>> I agree. It doesn't seem like that big of a task to move the rest >>>>>> of the stuff into the wizard (now you already have the >>>>>> framework). >>>>>> >>>>>> >>>>>> >>>>> Putting stuffs in the wizard goes against the packaging logic. On debian >>>>> you would want to use debconf to ask the user on how to configure his >>>>> node... >>>>> >>>>> >>>>> >>>> Both ways should probably be supported. >>>> >>>> >>> I see... and how exactly is that going to reduce the maintainance cost >>> again? I must have missed something. >>> >>> >> It's more about "doing things right", really. Obviously that does >> require extra work in the beginning, but if you look at the big picture, >> I think you will save time in the end by unifying the installation >> procedure with the other software in the world. I won't try to convince >> you about that if you disagree, because that would kind of be even more >> a waste of time. >> >> > > You don't get the picture: we are facing a human^wcodemonkey resource > shortage here. There is no need to convince me that each packaging > system has its assets and that having a properly written, maintained > package for each of them would be great... I am convinced of that... But > I do know that they are other areas of the code that needs improving > too. Not to mention that packaging (and more globally speaking dealing > with installation/platform specific issues) is everything *but* fun. > That's why I am not spending any more of my time on those related > matters. > > Right now we do provide and maintain *one* cross-platform installer; It's > arguably not perfect (hehe...) and does have defaults. It could use some > improvements... but no one is willing to work on that. > > I do not think that it's realistic for the project to spend any time > attempting to provide packages for any platform. Really that job needs > to be outsourced. I am one of those who think we shouldn't even provide > binary builds but source code. Many projects do it this way... and they > do find people to build and package their code. > > I will write some documentation on how to build and distribute freenet > in a cross-platform way. Then it will be up to people to follow the > guidelines and package the software. My guess is that writing that > documentation will be a waste of my time... but well... Let's try to be > optimistic for once. > I hope time will prove you wrong then :) > So far no one has even replied to Ian's call for help > (http://archives.freenetproject.org/message/20081124.201152.ab6aeaba.en.html). > I already opted in for trying to find time for it. Guess that's the best you get for now :P - Zero3
[freenet-dev] Installation -
Florent Daigni?re skrev: > * Zero3 [2008-11-26 22:39:58]: > > >>>> Will there be one available within reasonable time perhaps, or will >>>> we have to depend on the non-free one later on? >>>> >>> Ensuring that the code works reliably on other jvms takes dev's time >>> we'd rather spare somewhere else. It's all a matter of priorities, like >>> usual. >>> >>> >> We all know? That didn't quite answer though: >> >> >>>> Will there be one available within reasonable time perhaps, >>>> > > Go and ask people who write free jvms, not me. > Could be that you had some knowledge about the "jvm market", since you know that they at least do not provide the features Freenet needs? > >> and >> >> >>>> will we have to depend on the non-free one later on? >>>> > > Why would we have to depend on something? To get our packages included > into the main repository of some distributions (that includes debian), > yes, we would have to get rid of our dependancy on non-free software. > > But that's not our problem: that's the packager's one, isn't it? There > is no reason to make it *our* problem. > > You said we did, not me? Surely it is the devs problem if Freenet can't be made available in linux distro repositories because of depencies to non-free software? >>>>>>> The idea is to minimize the amount of data to download in order >>>>>>> to both spare bandwidth and reduce the overall installation >>>>>>> time. >>>>>>> >>>>>>> >>>>>> Not worth the trouble/annoyances/extra download time/... IMHO. >>>>>> >>>>>> >>>>> That's your view, not mine. Come back with figures and real arguments if >>>>> you plan to be convincing. Last time I checked I am the one who wrote >>>>> that part of the code... So I am the one who decides how it's done. >>>>> >>>>> >>>>> >>>> That seems like an awfully closed-minded attitude for a collaborative >>>> open-source project like Freenet. >>>> >>>> Being hosted at SourceForge, I can't see bandwidth being a problem? >>>> >>>> >>>> >>> We left SourceForge years ago because of their chronical unreliability. >>> >>> >> Oh. What's http://sourceforge.net/projects/freenet/ all about then? >> >> > > Nothing.. PR I guess... I told toad to get rid of it. Did you see that > there is only a README downloadable over there? > > Look closer: http://sourceforge.net/project/showfiles.php?group_id=978&package_id=973 >>>> But since you want some figures: I just did a test install. >>>> Downloading and setting up the plugins took the installer ~10 >>>> seconds on a 2 year old mainstream laptop with Windows XP. The >>>> plugins take up 383 KB. I don't know how many people that uncheck >>>> any or all of the plugins before installing, but I doubt it's a >>>> large part. Even if *everybody* unchecked all plugins in the >>>> installer and we assume nobody will ever install them later on, the >>>> overhead would be less than 4% of the ~10 MB that was downloaded >>>> during the install. In reality, that number will be *much* smaller >>>> as many people *will* install the plugins. If SourceForge can't >>>> keep up with that little extra bandwidth, I'll be glad to donate. >>>> >>>> >>>> >>> We did call for mirrors a while back, and we usually do before we >>> announce any new release. >>> >>> Right now we have 6 working ones and 13 configured. >>> >>> >>> >> What are the requirements, besides standard HTTP access to the actual files? >> >> > > Being able to run a gnu/rsync client to get the files on a regular > basis. We do ask mirrors to pull updates very frequently. > > I've got an account on dreamhost.net with virtually unlimited bandwidth and shell/cron access, so if you can provide me with whatever command line arguments rsync should use, it shouldn't be a problem. It's not enterprise-level uptime, but I haven
[freenet-dev] Laptops still suck was Re: Installation -
Florent Daigni?re skrev: > * Zero3 [2008-11-26 22:49:14]: > > >> Ian Clarke skrev: >> >>> On Wed, Nov 26, 2008 at 3:01 PM, Zero3 wrote: >>> >>> >>>>> The aim was to reduce the number of questions we ask during the >>>>> installation to a minimum: on the basis that advanced users can change >>>>> the settings they need afterwards, including whether the node >>>>> auto-starts or not. >>>>> >>>>> >>>>> >>>> Indeed, yet that is one of the questions you probably *ought* to ask. >>>> >>>> >>> I don't know - does mysql server ask this before it installs? I think >>> so long as there is an option to disable, we should default to >>> whatever is better for the network. The user has volunteered to run >>> Freenet after all. Remember also that Napster probably wouldn't have >>> worked if they hadn't defaulted to auto-running. Obviously Freenet != >>> Napster, but the analogy is valid in this case. >>> >>> >>> >> I get your point, but most servers with a default install don't sit in >> the background eating 200 MB of ram and a share of your download/upload. >> A default MySQL daemon is ~20 MB of ram and zero internet usage. >> > > Mysql is a server but not a peer to peer application. > > Exactly... >> I don't >> know about Napster, but IIRC it had a tray icon serving the purpose of >> both informing the user of the fact that it was unning, and allowing the >> user to easily kill it if needed. >> >> > > Writing a tray icon has been on the todo for a while; there is some code > for it in svn: it's called blueBunny. > > Finish it if you think it's important. > Erm... I was arguing about why autorunning Freenet might not work in the same way as it did with Napster, because of the missing tray icon. I wasn't complaining about the fact that it is missing. So no need to shout "do it yourself" all the time, really. I think I realized your feelings about suggestions after the last 5 times you responded in that way already. >> I do understand that whether the default should be on or off is a tough >> decision, but shouldn't we at least warn the user about it, and provide >> an easy way to disable it (if enabled by default)? >> >> > > Huh. > > On windows we use services which is the standard way of interfacing with > the OS... and on *nix we provide a start/stop script which is compatible > with init's format. What exactly isn't "easy" here? > > I'm quite sure the average Windows user doesn't know how to disable a system service (after he manages to figure out that Freenet has installed itself as a system service...) . The average geekiness among Windows users is much lower than among Linux users, mind you. Obviously, I cannot provide any raw numbers as you probably would prefer, so I guess we won't be getting any further on this matter either. - Zero3
